Python

1. What exactly are pandas? 

Pandas is a Python package that provides a large number of data structures for data-driven activities. Pandas fit in any function of data operation, whether it’s academics or tackling complicated corporate challenges, thanks to their interesting characteristics. Pandas is one of the most important programs to master because it can handle a wide range of files.

2. What are data frames, exactly?

A changeable data structure in pandas is called a data frame. Pandas can handle data that is organized on two axes and is heterogeneous. (Columns and rows)

Using pandas to read files:

– \s1 Import pandas using the PD df=p command. read CSV(“mydata.csv”)

df is a pandas data frame in this case. In pandas, read CSV() is used to read a comma-separated file as a data frame.

3. What is a Pandas Series, and what does it entail?

A series is a one-dimensional data structure in Pandas that may hold data of nearly any type. It has the appearance of an excel column. It is used for single-dimensional data manipulations and supports multiple operations.

4. What does it mean when pandas form a group?

A pandas groupby is a feature that allows you to separate and group objects using pandas. It is used to group data by classes, entities, which could then be used for aggregation, similar to SQL/MySQL/oracle group by. One or more columns can be used to group a data frame.

5. How do I make a data frame out of a list?

To make a data frame from a list,

1) start by making an empty data frame.

2)Add lists to the list as individual columns.

6. What is the best way to make a data frame from a dictionary?

To generate a data frame, a dictionary can be explicitly supplied as an input to the DataFrame() function.

7. In Pandas, how do you mix data frames?

The concat(), append(), and join() functions in pandas can stack two separate data frames horizontally or vertically.

Concat is a vertical stacking of data frames into a single data frame that works best when the data frames have the same columns and can be used for concatenation of data with comparable fields.

Append() is used to stack data frames horizontally. This is the finest concatenation function to use when merging two tables (data frames).

When we need to extract data from many data frames with one or more common columns, we utilize join. In this situation, the stacking is horizontal.

8. What types of joins can Panda provide?

A left join, an inner join, a right join, and an outside join are all present in Pandas.

9. In Pandas, how do you merge data frames?

The type and fields of the different data frames being merged determine how they are combined. If the data has identical fields, it is combined along axis 0, otherwise, it is merged along axis 1.

10. What is the best way to get the first five entries of a data frame?

We may get the top five entries of a data frame using the head(5) function. df.head() returns the top 5 rows by default. df.head(n) will be used to fetch the top n rows.

11. How can I get to a data frame’s last five entries?

We may get the top five entries of a data frame using the tail(5) method. df.tail() returns the top 5 rows by default. df.tail(n) will be used to fetch the last n rows.

12. What are comments, and how do you add them to a Python program?

A piece of text meant for information is referred to as a comment in Python. It’s especially important when multiple people are working on a set of codes. It can be used to inspect code, provide feedback, and troubleshoot it. There are two different categories of comments:

1. Comment on a single line
2. Comment with many lines

13. In Python, what is the difference between a list and a tuple?

Tuples are immutable, whereas lists are mutable.

14. In Python, what is a dictionary? Give a specific example.

A Python dictionary is a list of elements that are not in any particular order. Keys and values are written in curly brackets in Python dictionaries. The retrieval of value for known keys is optimized in dictionaries.

15. Find out the mean, median and standard deviation of this numpy array -> np.array([1,5,3,100,4,48])

import numpy as np n1=np.array([10,20,30,40,50,60]) print(np.mean(n1)) print(np.median(n1)) print(np.std(n1))

16. What is the definition of a classifier?

Any data point’s class is predicted using a classifier. Classifiers are hypotheses that are used to assign labels to data items based on their classification. To understand the relationship between input variables and the class, a classifier frequently needs training data. In Machine Learning, classification is a supervised learning strategy.

Any data point’s class is predicted using a classifier. Classifiers are hypotheses that are used to assign labels to data items based on their classification.

17. How do you change a string to lowercase in Python?

The method: can be used to convert all uppercase characters in a string to lowercase characters.

string.lower()

ex: string = ‘GREATLEARNING’ print(string.lower())

o/p: greatlearning

18. What’s the best way to get a list of all the keys in a dictionary?

We can get a list of keys in a variety of ways, including: dict.keys()

This method retrieves all of the dictionary’s accessible keys.

dict = {1:a, 2:b, 3:c} dict.keys()

o/p: [1, 2, 3]

19. How do you capitalize a string’s first letter?

To capitalize the initial character of a string, we can use the capitalize() method. If the initial character is already capitalized, the original string is returned.

Syntax: string_name.capitalize() ex: n = “greatlearning” print(n.capitalize())

o/p: Greatlearning

20. In Python, how do you insert an element at a specific index?

The insert() function is a built-in Python function. It’s possible to use it to insert an element at a specific index.

Syntax: list_name.insert(index, element)

ex: list = [ 0,1, 2, 3, 4, 5, 6, 7 ]

#insert 10 at 6th index

list.insert(6, 10)

o/p: [0,1,2,3,4,5,10,6,7]

21. How are you going to get rid of duplicate elements from a list?

To delete duplicate elements from a list, you can use a variety of techniques. The most typical method is to use the set() function to convert a list into a set, then use the list() function to convert it back to a list if necessary.

22. What exactly is recursion?

A recursive function is one that calls itself one or more times within its body. One of the most significant requirements for using a recursive function in a program is that it must end, otherwise, an infinite loop would occur.

23. Explain how to use Python’s List Comprehension feature.

List comprehensions are used to change one list into a different one. Elements can be included in the new list on a conditional basis, and each member can be modified as needed. It consists of a bracketed statement that precedes a for clause.

24. What is the purpose of the bytes() function?

A bytes object is returned by the bytes() function. It’s used to convert things to bytes objects or to produce empty bytes objects of a given size.

25. What are the various types of Python operators?

The following are the basic operators in Python:

Arithmetic ( +, -, Multiplication(*), Division(/), Modulus( percent ) ), Relational (, >, =, >=, ==,!=, ), Assignment ( =. +=, -=, /=, *=, percent = ), Logical ( =. +=, -=, /=, *=, percent = ), ( and, or not ), Bitwise Operators, Membership, and Identity

26. What exactly is a ‘with statement’?

In Python, the “with” statement is used to handle exceptions. Without utilizing the close() function, a file can be opened and closed while executing a block of code containing the “with” line. It basically makes the code a lot easier to read.

27. In Python, what is the map() function?

In Python, the map() method is used to apply a function to all components of an iterable. Function and iterable are the two parameters that make up this function. The function is supplied as an argument, and it is then applied to all elements of an iterable (which is passed as the second parameter). As a result, an object list is returned.

28. In Python, what is __init__?

In Python, the _init_ function, often known as the function Object() { [native code] } in OOP, is a reserved method. When a class is used to construct an object, the _init_ method is used to access the class attributes.

29. What tools are available to perform the static analysis?

Pychecker and Pylint are two static analysis tools for finding flaws in Python. Pychecker finds flaws in source code and issues warnings regarding the style and complexity of the code. Pylint, on the other hand, checks whether the module adheres to a coding standard.

30. What is the difference between tuple and dictionary?

A tuple differs from a dictionary in that a dictionary is mutable, whereas a tuple is not. In other words, a dictionary’s content can be modified without affecting its identity, but this is not allowed with a tuple.

Dictionary is one of Python’s built-in datatypes. It establishes a one-to-one correspondence between keys and values. Dictionary keys and values are stored in pairs in dictionaries. Keys are used to index dictionaries.

31. In Python, what is the meaning of pass?

Pass is a statement that has no effect when used. To put it another way, it’s a Null statement. The interpreter does not ignore this statement, but no action is taken as a result of it. It’s used when you don’t want any commands to run but yet need to make a statement.

The pass statement is used when there’s a syntactic but not an operational requirement. For example – The program below prints a string ignoring the spaces.

var=”Si mplilea rn”

for i in var:

if i==” “:

pass

else:

print(i,end=””)

Here, the pass statement refers to ‘no action required.’

32. In Python, how do you copy an object?

Although not all objects can be duplicated in Python, the majority of them can. To copy an object to a variable, we can use the “=” operator.

33. How do you turn a number into a string?

To convert a number to a string, use the built-in function str().

34. What are the differences between a module and a package in Python?

Modules are the building blocks of a program. A module is a Python software file that imports other characteristics and objects. A program’s folder is a collection of modules. Modules and subfolders can be found in a package.

35. In Python, what is the object() function?

The object() method in Python returns an empty object. This object can’t have any new attributes or methods added to it.

36. What do NumPy and SciPy have in common?

SciPy stands for Scientific Python, while NumPy stands for Numerical Python. NumPy is the basic library for defining arrays and solving elementary mathematical issues, whereas SciPy is used for more sophisticated problems like numerical integration, optimization, and machine learning.

37. What does len() do?

len() is used to determine the length of a string, a list, an array, and so on. ex: str = “greatlearning” print(len(str)) o/p: 13

38. What does encapsulation mean in Python?

Encapsulation refers to the joining of code and data. Consider a Python class.

39. In Python, what is the type ()?

type() is a built-in method that returns the object’s type or creates a new type object based on the inputs passed in.

40. What is the purpose of the split() function?

Split is a function that divides a string into shorter strings using defined separators.

The split() function splits a string into a number of strings based on a specific delimiter.

Syntax –

string.split(delimiter, max)

Where:

the delimiter is the character based on which the string is split. By default it is space.

max is the maximum number of splits

Example –

>>var=“Red,Blue,Green,Orange”

>>lst=var.split(“,”,2)

>>print(lst)

Output:

[‘Red’,’Blue’,’Green, Orange’]

Here, we have a variable var whose values are to be split with commas. Note that ‘2’ indicates that only the first two values will be split.

41. What are built-in types does python provide?

Python includes the following data types:

Numbers: Python distinguishes between three types of numbers:

1. All positive and negative numbers without a fractional part are integers.
2. Float: Any real number that may be represented in floating-point format.
3. Complex numbers: x+yj represents a number with a real and imaginary component. x and y are floats, and j is -1 (often known as an imaginary number because of its square root).

Boolean: The Boolean data type is a data type that can only have one of two values: True or False. The letters ‘T’ and ‘F’ are capitalized.

A string value is made up of one or more characters enclosed in single, double, or triple quotations.

List: A list object is an ordered collection of one or more data objects in square brackets, which might be of different types. Individual elements in a list can be added, edited, or deleted since they are modifiable.

Set: Curly brackets encompass an unordered group of unique objects.

Frozen set: They’re similar to sets, but they’re immutable, meaning we can’t change their values after they’ve been created.

Dictionary: A dictionary is an unordered object in which each item has a key and we may retrieve each value using that key. Curly brackets surround a collection of similar pairs.

42. In Python, how do you reverse a string?

There are no built-in functions in Python to let us reverse a string. For this, we’ll need to use an array slicing operation.

>> #Input

>>df = pd.DataFrame(np.arange(25).reshape(5, -1))

>> #Solution

>>df.iloc[::-1, :]

43. How do I find out the Python version in CMD?

Press CMD + Space to see the Python version in CMD. This activates Spotlight. Type “terminal” into this box and hit Enter. Type python –version or python -V and press enter to run the program. The python version will be returned in the line following the command.

44. When it comes to identifiers, is Python case sensitive?

Yes. When it comes to identifiers, Python is case-sensitive. It’s a case-by-case language. As a result, variable and Variable are not synonymous.

45. How can I use values from existing columns to build a new column in Pandas?

On a pandas data frame, we can conduct column-based mathematical operations. Operators can be used on Pandas columns that contain numeric values.

46. What are the many functions that grouby in pandas can perform?

Multiple aggregate functions can be utilised with grouby() in pandas. sum(), mean(), count(), and standard are a few examples().

Data is separated into groups based on categories, and the data in these individual groups can then be aggregated using the functions listed above.

47. In Pandas, how do you choose columns and add them to a new data frame? What if two columns with the same name exist?

If df is a pandas data frame, df.columns return a list of all columns. We may then select columns to create new columns.

If two columns have the same name, both of them are copied to the new data frame.

48. How to delete a column or group of columns in pandas?

drop() function can be used to delete the columns from a data frame.

49. Given the following data frame drop rows having column values as A.

	Col1	Col2
0	1	A
1	2	B
2	3	C

 

Code: d={“col1″:[1,2,3],”col2”:[“A”,”B”,”C”]} df=pd.DataFrame(d) df.dropna(inplace=True) df=df[df.col1!=1] df

Output:

	Col1	Col2
1	2	B
2	3	C

 

50. What is reindexing in pandas?

Reindexing is the process of re-assigning the index of a pandas data frame.

51. What exactly do you mean when you say “lambda function”?

A lambda function is a type of anonymous function. This function can take as many parameters as you want, but just one statement.

Lambda is typically utilized in instances where an anonymous function is required for a short period of time. Lambda functions can be applied in two different ways:

• Assigning Lambda functions to a variable
• Wrapping Lambda function into another function

Create a lambda function that prints the total of all the elements in this list -> [5, 8, 10, 20, 50, 100].

A lambda function is an anonymous function (a function that does not have a name) in Python. To define anonymous functions, we use the ‘lambda’ keyword instead of the ‘def’ keyword, hence the name ‘lambda function’. Lambda functions can have any number of arguments but only one statement.

<code>from functools import reduce sequences = [5, 8, 10, 20, 50, 100] sum = reduce

(lambda x, y: x+y, sequences) print(sum)</code>

52. What is vstack() in numpy? Give an example

vstack() is a function to align rows vertically. All rows must have the same number of elements.

53. How do we interpret Python?

When a python program is written, it converts the source code written by the developer into an intermediate language, which is then converted into machine language that needs to be executed.

54. How to remove spaces from a string in Python?

Spaces can be removed from a string in python by using strip() or replace() functions. Strip() function is used to remove the leading and trailing white spaces while the replace() function is used to remove all the white spaces in the string.

To Remove All Leading Whitespace in a String

Python provides the inbuilt function lstrip() to remove all leading spaces from a string.

>>“      Python”.lstrip

Output: Python

55. Explain the file processing modes that Python supports.

There are three file processing modes in Python: read-only(r), write-only(w), read-write(rw) and append (a). So, if you are opening a text file in say, read mode. The preceding modes become “rt” for read-only, “wt” for write, and so on. Similarly, a binary file can be opened by specifying “b” along with the file accessing flags (“r”, “w”, “rw”, and “a”) preceding it.

56. How is memory managed in Python?

Memory management in python comprises a private heap containing all objects and data structure. The heap is managed by the interpreter and the programmer does not have access to it at all. The Python memory manager does all the memory allocation. Moreover, there is an inbuilt garbage collector that recycles and frees memory for the heap space.

57. What is a unit test in Python?

Unit test is a unit testing framework in Python. It supports sharing of setup and shutdown code for tests, aggregation of tests into collections, test automation, and independence of the tests from the reporting framework.

58. How do you delete a file in Python?

Files can be deleted in Python by using the command os.remove (filename) or os.unlink(filename)

Use command os.remove(file_name) to delete a file in Python.

59. How do you create an empty class in Python?

To create an empty class we can use the pass command after the definition of the class object. A pass is a statement in Python that does nothing.

60. What are Python decorators?

Decorators are functions that take another function as an argument to modify its behavior without changing the function itself. These are useful when we want to dynamically increase the functionality of a function without changing it.

61. You have this covid-19 dataset below:

From this dataset, how will you make a bar-plot for the top 5 states having maximum confirmed cases as of 17=07-2020?

sol: #keeping only required columns df = df[[‘Date’, ‘State/UnionTerritory’,’Cured’,’Deaths’,’Confirmed’]] #renaming column names df.columns = [‘date’, ‘state’,’cured’,’deaths’,’confirmed’] #current date today = df[df.date == ‘2020-07-17’] #Sorting data w.r.t number of confirmed cases max_confirmed_cases=today.sort_values(by=”confirmed”,ascending=False) max_confirmed_cases #Getting states with maximum number of confirmed cases top_states_confirmed=max_confirmed_cases[0:5] #Making bar-plot for states with top confirmed cases sns.set(rc={‘figure.figsize’:(15,10)}) sns.barplot(x=”state”,y=”confirmed”,data=top_states_confirmed,hue=”state”) plt.show()

62. From this covid-19 dataset:

How can you make a bar plot for the top-5 states with the most amount of deaths?

 

Sol: max_death_cases=today.sort_values(by=”deaths”,ascending=False) max_death_cases sns.set(rc={‘figure.figsize’:(15,10)}) sns.barplot(x=”state”,y=”deaths”,data=top_states_death,hue=”state”) plt.show()

63. In Python, what is “self”?

Self is a class instance or an object. In Python, this is explicitly supplied as the first parameter. In Java, on the other hand, it is optional. With local variables, it makes it easier to distinguish between a class’s methods and attributes.

In the init method, the self variable refers to the newly created object, whereas it relates to the object whose method was called in other methods.

Self is used to represent the class instance. In Python, you can access the class’s attributes and methods with this keyword. It connects the attributes to the arguments. Self appears in a variety of contexts and is frequently mistaken for a term. Self is not a keyword in Python, unlike in C++.

64. What is the difference between the methods append() and extend()?

The methods append() and extend() are used to add elements to the end of a list.

append(element): Adds the given element at the end of the list that called this append() method

extend(another-list): Adds the elements of another list at the end of the list that called this extend() method

65. How does Python Flask handle database requests?

Flask supports a database-powered application (RDBS). Such a system requires creating a schema, which needs piping the schema.sql file into the sqlite3 command. Python developers need to install the sqlite3 command to create or initiate the database in Flask.

Flask allows to request for a database in three ways:

• before_request(): They are called before a request and pass no arguments.
• after_request(): They are called after a request and pass the response that will be sent to the client.
• teardown_request(): They are called in a situation when an exception is raised and responses are not guaranteed. They are called after the response has been constructed. They are not allowed to modify the request, and their values are ignored.

66. In Python, what exactly is docstring?

The string literals encased in triple quotes that come right after the definition of a function, method, class, or module are called Python docstrings. The functionality of a function, method, class, or module is typically described using these terms. Using the __doc__ attribute, we may retrieve these docstrings.

This is one of the most frequently asked Python interview questions

Docstrings are used in providing documentation to various Python modules, classes, functions, and methods.

Example –

def add(a,b):

” ” “This function adds two numbers.” ” ”

sum=a+b

return sum

sum=add(10,20)

print(“Accessing doctstring method 1:”,add.__doc__)

print(“Accessing doctstring method 2:”,end=””)

help(add)

Output –

Accessing docstring method 1: This function adds two numbers.

Accessing docstring method 2: Help on function add-in module __main__:

add(a, b)

This function adds two numbers.

Docstrings are documentation strings. Within triple quotations are these docstrings. They are not allocated to any variable and, as a result, they can also be used as comments.

67. How is Multithreading achieved in Python?

Python has a multi-threading package, but commonly not considered good practice to use it as it will result in increased code execution time.

• Python has a constructor called the Global Interpreter Lock (GIL). The GIL ensures that only one of your ‘threads’ can execute at one time. The process makes sure that a thread acquires the GIL, does a little work, then passes the GIL onto the next thread.
• This happens at a very Quick instance of time and that’s why to the human eye it seems like your threads are executing parallelly, but in reality, they are executing one by one by just taking turns using the same CPU core.

Multithreading usually implies that multiple threads are executed concurrently. The Python Global Interpreter Lock doesn’t allow more than one thread to hold the Python interpreter at that particular point of time. So multithreading in python is achieved through context switching. It is quite different from multiprocessing which actually opens up multiple processes across multiple threads.

Although Python includes a multi-threading module, it is usually not a good idea to utilize it if you want to multi-thread to speed up your code.

As this happens so quickly, it may appear to the human eye that your threads are running in parallel, but they are actually sharing the same CPU core.

The Global Interpreter Lock is a Python concept (GIL). Only one of your ‘threads’ can execute at a moment, thanks to the GIL. A thread obtains the GIL, performs some work, and then passes the GIL to the following thread.

68. What is slicing in Python?

Slicing is a process used to select a range of elements from sequence data types like list, string, and tuple. Slicing is beneficial and easy to extract out the elements. It requires a : (colon) which separates the start index and end index of the field. All the data sequence types List or tuple allows us to use slicing to get the needed elements. Although we can get elements by specifying an index, we get only a single element whereas using slicing we can get a group or appropriate range of needed elements.

69. What is functional programming? Does Python follow a functional programming style? If yes, list a few methods to implement functionally oriented programming in Python.

Functional programming is a coding style where the main source of logic in a program comes from functions.

Incorporating functional programming in our codes means writing pure functions.

Pure functions are functions that cause little or no changes outside the scope of the function. These changes are referred to as side effects. To reduce side effects, pure functions are used, which makes the code easy to follow, test, or debug.

70. Which one of the following is not the correct syntax for creating a set in Python?

1. set([[1,2],[3,4],[4,5]])
2. set([1,2,2,3,4,5])
3. {1,2,3,4}
4. set((1,2,3,4))

set([[1,2],[3,4],[4,5]])

Explanation: The argument given for the set must be iterable.

71. What is the difference between / and // operator in Python?

• /: is a division operator and returns the Quotient value.

10/3

3.33

• //: is known as floor division operator and used to return only the value of quotient before the decimal

10//3

3

72. What is the easiest way to calculate percentiles when using Python?

The easiest and the most efficient way you can calculate percentiles in Python is to make use of NumPy arrays and its functions.

73. What is a palindrome number?

A palindrome is a word, phrase, or sequence that reads the same forward as it does backward, such as madam, nurses run, and so on.

74. What is slicing in python?

Slicing is used to access parts of sequences like lists, tuples, and strings. The syntax of slicing is -[start:end:step]. The step can be omitted as well. When we write [start:end] this returns all the elements of the sequence from the start (inclusive) till the end-1 element. If the start or end element is negative i, it means the ith element from the end. The step indicates the jump or how many elements have to be skipped. Eg. if there is a list- [1,2,3,4,5,6,7,8]. Then [-1:2:2] will return elements starting from the last element till the third element by printing every second element.i.e. [8,6,4].

75. What are Keywords in Python?

Keywords in python are reserved words that have special meanings. They are generally used to define types of variables. Keywords cannot be used for variable or function names. There are the following 33 keywords in python –

• And
• Or
• Not
• If
• Elif
• Else
• For
• While
• Break
•  As
• Def
• Lambda
• Pass
• Return
• True
• False
• Try
• With
• Assert
• Class
• Continue
• Del
• Except
• Finally
• From
• Global
• Import
• In
• Is
• None
• Nonlocal
• Raise
• Yield

 

76. What are the new features added in Python 3.9.0.0 version?

The new features in Python 3.9.0.0 version are-

• New Dictionary functions Merge(|) and Update(|=)
• New String Methods to Remove Prefixes and Suffixes
• Type Hinting Generics in Standard Collections
• New Parser based on PEG rather than LL1
• New modules like zoneinfo and graphlib
• Improved Modules like ast, asyncio, etc.
• Optimizations such as optimized idiom for assignment, signal handling, optimized python built ins, etc.
• Deprecated functions and commands such as deprecated parser and symbol modules, deprecated functions, etc.
• Removal of erroneous methods, functions, etc.

 

77. How is memory managed in Python?

Memory is managed in Python in the following ways:

Memory management in python is managed by Python private heap space. All Python objects and data structures are located in a private heap. The programmer does not have access to this private heap. The python interpreter takes care of this instead.

The allocation of heap space for Python objects is done by Python’s memory manager. The core API gives access to some tools for the programmer to code.

Python also has an inbuilt garbage collector, which recycles all the unused memory and so that it can be made available to the heap space.

Python has a private heap space that stores all the objects. The Python memory manager regulates various aspects of this heap, such as sharing, caching, segmentation, and allocation. The user has no control over the heap; only the Python interpreter has access.

• Python’s private heap space is in charge of memory management. A private heap holds all Python objects and data structures. This secret heap is not accessible to the programmer. Instead, the Python interpreter takes care of it.
• Python also includes a built-in garbage collector, which recycles all unused memory and makes it available to the heap space.
• Python’s memory management is in charge of allocating heap space for Python objects. The core API allows programmers access to some programming tools.

78. What is namespace in Python?

A namespace is a naming system used to make sure that names are unique to avoid naming conflicts.

79. What is PYTHONPATH?

It is an environment variable that is used when a module is imported. Whenever a module is imported, PYTHONPATH is also looked up to check for the presence of the imported modules in various directories. The interpreter uses it to determine which module to load.

80. What are python modules? Name some commonly used built-in modules in Python?

Python modules are files containing Python code. This code can either be functions classes or variables. A Python module is a .py file containing executable code.

Some of the commonly used built-in modules are:

• os
• sys
• math
• random
• data time
• JSON

 

81. How to remove values to a python array?

Elements can be removed from the python array using pop() or remove() methods.

pop(): This function will return the removed element.

remove(): It will not return the removed element.

82. What are Python libraries? Name a few of them.

Python libraries are a collection of Python packages. Some of the majorly used python libraries are – Numpy, Pandas, Matplotlib, Scikit-learn, and many more.

83. How do you do data abstraction in Python?

Data Abstraction is providing only the required details and hiding the implementation from the world. It can be achieved in Python by using interfaces and abstract classes.

84. What does an object() do?

It returns a featureless object that is a base for all classes. Also, it does not take any parameters.

85. What is the Difference Between a Shallow Copy and Deep Copy?

Deepcopy creates a different object and populates it with the child objects of the original object. Therefore, changes in the original object are not reflected in the copy.

copy.deepcopy() creates a Deep Copy.

Shallow copy creates a different object and populates it with the references of the child objects within the original object. Therefore, changes in the original object are reflected in the copy.

copy.copy creates a Shallow Copy.

86. What Advantage Does the Numpy Array Have over a Nested List?

Numpy is written in C so that all its complexities are backed into a simple to use a module. Lists, on the other hand, are dynamically typed. Therefore, Python must check the data type of each element every time it uses it. This makes Numpy arrays much faster than lists.

Numpy has a lot of additional functionality that list doesn’t offer; for instance, a lot of things can be automated in Numpy.

87. What are Pickling and Unpickling?

 

Pickling	Unpickling
Converting a Python object hierarchy to a byte stream is called pickling Pickling is also referred to as serialization	Converting a byte stream to a Python object hierarchy is called unpickling Unpickling is also referred to as deserialization

If you just created a neural network model, you can save that model to your hard drive, pickle it, and then unpickle to bring it back into another software program or to use it at a later time.

Pickling is the process of converting a Python object hierarchy into a byte stream for storing it into a database. It is also known as serialization. Unpickling is the reverse of pickling. The byte stream is converted back into an object hierarchy.

The Pickle module takes any Python object and converts it to a string representation, which it then dumps into a file using the dump method. This is known as pickling. Unpickling is the process of recovering original Python objects from a stored text representation.

88. Are Arguments in Python Passed by Value or by Reference?

Arguments are passed in python by a reference. This means that any changes made within a function are reflected in the original object.

Consider two sets of code shown below:

In the first example, we only assigned a value to one element of ‘l’, so the output is [3, 2, 3, 4].

In the second example, we have created a whole new object for ‘l’. But, the values [3, 2, 3, 4] doesn’t show up in the output as it is outside the definition of the function.

89. How Would You Generate Random Numbers in Python?

To generate random numbers in Python, you must first import the random module.

The random() function generates a random float value between 0 & 1.

> random.random()

The randrange() function generates a random number within a given range.

Syntax: randrange(beginning, end, step)

Example – > random.randrange(1,10,2)

90. What Does the // Operator Do?

In Python, the / operator performs division and returns the quotient in the float.

For example: 5 / 2 returns 2.5

The // operator, on the other hand, returns the quotient in integer.

For example: 5 // 2 returns 2

91. What Does the ‘is’ Operator Do?

The ‘is’ operator compares the id of the two objects.

list1=[1,2,3]

list2=[1,2,3]

list3=list1

list1 == list2 🡪 True

list1 is list2 🡪 False

list1 is list3 🡪 True

92. How Will You Check If All the Characters in a String Are Alphanumeric?

Python has an inbuilt method isalnum() which returns true if all characters in the string are alphanumeric.

Example –

>> “abcd123”.isalnum()

Output: True

>>”abcd@123#”.isalnum()

Output: False

Another way is to use regex as shown.

>>import re

>>bool(re.match(‘[A-Za-z0-9]+$’,’abcd123’))

Output: True

>> bool(re.match(‘[A-Za-z0-9]+$’,’abcd@123’))

Output: False

93. How Will You Merge Elements in a Sequence?

There are three types of sequences in Python:

• Lists
• Tuples
• Strings

Example of Lists –

>>l1=[1,2,3]

>>l2=[4,5,6]

>>l1+l2

Output: [1,2,3,4,5,6]

Example of Tuples –

>>t1=(1,2,3)

>>t2=(4,5,6)

>>t1+t2

Output: (1,2,3,4,5,6)

Example of String –

>>s1=“Simpli”

>>s2=“learn”

>>s1+s2

Output: ‘Simplilearn’

94. How Would You Replace All Occurrences of a Substring with a New String?

The replace() function can be used with strings for replacing a substring with a given string. Syntax:

str.replace(old, new, count)

replace() returns a new string without modifying the original string.

Example –

>>”Hey John. How are you, John?”.replace(“john”,“John”,1)

Output: “Hey John. How are you, John?

95.What Is the Difference Between Del and Remove() on Lists?

del	remove()
del removes all elements of a list within a given range Syntax: del list[start:end]	remove() removes the first occurrence of a particular character Syntax: list.remove(element)

Here is an example to understand the two statements –

>>lis=[‘a’, ‘b’, ‘c’, ‘d’]

>>del lis[1:3]

>>lis

Output: [“a”,”d”]

>>lis=[‘a’, ‘b’, ‘b’, ‘d’]

>>lis.remove(‘b’)

>>lis

Output: [‘a’, ‘b’, ‘d’]

Note that in the range 1:3, the elements are counted up to 2 and not 3.

96. How Do You Display the Contents of a Text File in Reverse Order?

You can display the contents of a text file in reverse order using the following steps:

• Open the file using the open() function
• Store the contents of the file into a list
• Reverse the contents of the list
• Run a for loop to iterate through the list

 

 

97. Differentiate Between append() and extend().

append()	extend()
append() adds an element to the end of the list Example – >>lst=[1,2,3] >>lst.append(4) >>lst Output:[1,2,3,4]	extend() adds elements from an iterable to the end of the list Example – >>lst=[1,2,3] >>lst.extend([4,5,6]) >>lst Output:[1,2,3,4,5,6]

98. What Is the Output of the below Code? Justify Your Answer.

>>def addToList(val, list=[]):

>> list.append(val)

>> return list

>>list1 = addToList(1)

>>list2 = addToList(123,[])

>>list3 = addToList(‘a’)

>>print (“list1 = %s” % list1)

>>print (“list2 = %s” % list2)

>>print (“list3 = %s” % list3)

Output:

list1 = [1,’a’]

list2 = [123]

lilst3 = [1,’a’]

Note that list1 and list3 are equal. When we passed the information to the addToList, we did it without a second value. If we don’t have an empty list as the second value, it will start off with an empty list, which we then append. For list2, we appended the value to an empty list, so its value becomes [123].

For list3, we’re adding ‘a’ to the list. Because we didn’t designate the list, it is a shared value. It means the list doesn’t reset and we get its value as [1, ‘a’].

Remember that a default list is created only once during the function and not during its call number.

99. What Is the Difference Between a List and a Tuple?

Lists are mutable while tuples are immutable.

Example:

List

>>lst = [1,2,3]

>>lst[2] = 4

>>lst

Output:[1,2,4]

Tuple

>>tpl = (1,2,3)

>>tpl[2] = 4

>>tpl

Output:TypeError: ‘tuple’

the object does not support item

assignment

There is an error because you can’t change the tuple 1 2 3 into 1 2 4. You have to completely reassign tuple to a new value.

100. How Do You Use Print() Without the Newline?

The solution to this depends on the Python version you are using.

Python v2

>>print(“Hi. ”),

>>print(“How are you?”)

Output: Hi. How are you?

Python v3

>>print(“Hi”,end=“ ”)

>>print(“How are you?”)

Output: Hi. How are you?

101.Is Python Object-oriented or Functional Programming?

Python is considered a multi-paradigm language.

Python follows the object-oriented paradigm

• Python allows the creation of objects and their manipulation through specific methods
• It supports most of the features of OOPS such as inheritance and polymorphism

Python follows the functional programming paradigm

• Functions may be used as the first-class object
• Python supports Lambda functions which are characteristic of the functional paradigm

102. Write a Function Prototype That Takes a Variable Number of Arguments.

The function prototype is as follows:

def function_name(*list)

>>def fun(*var):

>> for i in var:

print(i)

>>fun(1)

>>fun(1,25,6)

In the above code, * indicates that there are multiple arguments of a variable.

103. What Are *args and *kwargs?

*args

• It is used in a function prototype to accept a varying number of arguments.
• It’s an iterable object.
• Usage – def fun(*args)
• The function definition uses the *args syntax to pass variable-length parameters.
• “*” denotes variable length, while “args” is the standard name. Any other will suffice.

 

*kwargs

• It is used in a function prototype to accept the varying number of keyworded arguments.
• It’s an iterable object
• Usage – def fun(**kwargs):
• **kwargs is a special syntax for passing variable-length keyworded arguments to functions.
• When a variable is passed to a function, it is called a keyworded argument.
• “Kwargs” is also used by convention here. You are free to use any other name.

 

fun(colour=”red”.units=2)

104. “in Python, Functions Are First-class Objects.” What Do You Infer from This?

It means that a function can be treated just like an object. You can assign them to variables, or pass them as arguments to other functions. You can even return them from other functions.

105. What Is the Output Of: Print(__name__)? Justify Your Answer.

__name__ is a special variable that holds the name of the current module. Program execution starts from main or code with 0 indentations. Thus, __name__ has a value __main__ in the above case. If the file is imported from another module, __name__ holds the name of this module.

106. What Is a Numpy Array?

A numpy array is a grid of values, all of the same type, and is indexed by a tuple of non-negative integers. The number of dimensions determines the rank of the array. The shape of an array is a tuple of integers giving the size of the array along each dimension.

107. What Is the Difference Between Matrices and Arrays?

Matrices	Arrays
A matrix comes from linear algebra and is a two-dimensional representation of data It comes with a powerful set of mathematical operations that allow you to manipulate the data in interesting ways	An array is a sequence of objects of similar data type An array within another array forms a matrix

108. How Do You Get Indices of N Maximum Values in a Numpy Array?

>>import numpy as np

>>arr=np.array([1, 3, 2, 4, 5])

>print(arr.argsort( ) [ -N: ][: : -1])

109. How Would You Obtain the Res_set from the Train_set and the Test_set from Below?

>>train_set=np.array([1, 2, 3])

>>test_set=np.array([[0, 1, 2], [1, 2, 3])

Res_set 🡪 [[1, 2, 3], [0, 1, 2], [1, 2, 3]]

Choose the correct option:

1. res_set = train_set.append(test_set)
2. res_set = np.concatenate([train_set, test_set]))
3. resulting_set = np.vstack([train_set, test_set])
4. None of these

Here, options a and b would both do horizontal stacking, but we want vertical stacking. So, option c is the right statement.

resulting_set = np.vstack([train_set, test_set])

110. You Have Uploaded the Dataset in Csv Format on Google Spreadsheet and Shared It Publicly. How Can You Access This in Python?

We can use the following code:

>>link = https://docs.google.com/spreadsheets/d/…

>>source = StringIO.StringIO(requests.get(link).content))

>data = pd.read_csv(source)

111. What Is the Difference Between the Two Data Series given Below?

df[‘Name’] and df.loc[:, ‘Name’], where:

df = pd.DataFrame([‘aa’, ‘bb’, ‘xx’, ‘uu’], [21, 16, 50, 33], columns = [‘Name’, ‘Age’])

Choose the correct option:

1. 1 is the view of original dataframe and 2 is a copy of original dataframe
2. 2 is the view of original dataframe and 1 is a copy of original dataframe
3. Both are copies of original dataframe
4. Both are views of original dataframe

Answer – 3. Both are copies of the original dataframe.

112. You Get the Error “temp.Csv” While Trying to Read a File Using Pandas. Which of the Following Could Correct It?

Error:

Traceback (most recent call last): File “<input>”, line 1, in<module> UnicodeEncodeError:

‘ascii’ codec can’t encode character.

Choose the correct option:

1. pd.read_csv(“temp.csv”, compression=’gzip’)
2. pd.read_csv(“temp.csv”, dialect=’str’)
3. pd.read_csv(“temp.csv”, encoding=’utf-8′)
4. None of these

The error relates to the difference between utf-8 coding and a Unicode.

So option 3. pd.read_csv(“temp.csv”, encoding=’utf-8′) can correct it.

113. How Do You Set a Line Width in the Plot given Below?

>>import matplotlib.pyplot as plt

>>plt.plot([1,2,3,4])

>>plt.show()

Choose the correct option:

1. In line two, write plt.plot([1,2,3,4], width=3)
2. In line two, write plt.plot([1,2,3,4], line_width=3
3. In line two, write plt.plot([1,2,3,4], lw=3)
4. None of these

Answer – 3. In line two, write plt.plot([1,2,3,4], lw=3)

114. How Would You Reset the Index of a Dataframe to a given List? Choose the Correct Option.

1. df.reset_index(new_index,)
2. df.reindex(new_index,)
3. df.reindex_like(new_index,)
4. None of these

Answer – 3. df.reindex_like(new_index,)

115. How Can You Copy Objects in Python?

The function used to copy objects in Python are:

copy.copy for shallow copy and

copy.deepcopy() for deep copy

The assignment statement (= operator) in Python does not copy objects. Instead, it establishes a connection between the existing object and the name of the target variable. The copy module is used to make copies of an object in Python. Furthermore, the copy module provides two options for producing copies of a given object –

Deep Copy: Deep Copy recursively replicates all values from source to destination object, including the objects referenced by the source object.

from copy import copy, deepcopy

list_1 = [1, 2, [3, 5], 4]

## shallow copy

list_2 = copy(list_1)

list_2[3] = 7

list_2[2].append(6)

list_2    # output => [1, 2, [3, 5, 6], 7]

list_1    # output => [1, 2, [3, 5, 6], 4]

## deep copy

list_3 = deepcopy(list_1)

list_3[3] = 8

list_3[2].append(7)

list_3    # output => [1, 2, [3, 5, 6, 7], 8]

list_1    # output => [1, 2, [3, 5, 6], 4]

Shallow Copy: A bit-wise copy of an object is called a shallow copy. The values in the copied object are identical to those in the original object. If one of the values is a reference to another object, only its reference addresses are copied.

When a new instance type is formed, a shallow copy is used to maintain the values that were copied in the previous instance. Shallow copy is used to copy reference pointers in the same way as values are copied. These references refer to the original objects, and any modifications made to any member of the class will have an impact on the original copy. Shallow copy enables faster program execution and is dependent on the size of the data being utilized.

Deep copy is a technique for storing previously copied values. The reference pointers to the objects are not copied during deep copy. It creates a reference to an object and stores the new object that is referenced to by another object. The changes made to the original copy will have no effect on any subsequent copies that utilize the item. Deep copy slows down program performance by creating many copies of each object that is called.

116. What Is the Difference Between range() and xrange() Functions in Python?

In terms of functionality, xrange and range are essentially the same. They both provide you the option of generating a list of integers to use whatever you want. The sole difference between range and xrange is that range produces a Python list object whereas x range returns an xrange object. This is especially true if you are working with a machine that requires a lot of memory, such as a phone because range will utilize as much memory as it can to generate your array of numbers, which can cause a memory error and crash your program. It is a beast with a memory problem.

 

range()	xrange()
range returns a Python list object	xrange returns an xrange object

117. How Can You Check Whether a Pandas Dataframe Is Empty or Not?

The attribute df.empty is used to check whether a pandas data frame is empty or not.

>>import pandas as pd

>>df=pd.DataFrame({A:[]})

>>df.empty

Output: True

118. Write a Code to Sort an Array in Numpy by the (N-1)Th Column.

This can be achieved by using argsort() function. Let us take an array X; the code to sort the (n-1)th column will be x[x [: n-2].argsoft()]

The code is as shown below:

>>import numpy as np

>>X=np.array([[1,2,3],[0,5,2],[2,3,4]])

>>X[X[:,1].argsort()]

Output:array([[1,2,3],[0,5,2],[2,3,4]])

119. How Do You Create a Series from a List, Numpy Array, and Dictionary?

The code is as shown:

>> #Input

>>import numpy as np

>>import pandas as pd

>>mylist = list(‘abcedfghijklmnopqrstuvwxyz’)

>>myarr = np.arange(26)

>>mydict = dict(zip(mylist, myarr))

>> #Solution

>>ser1 = pd.Series(mylist)

>>ser2 = pd.Series(myarr)

>>ser3 = pd.Series(mydict)

>print(ser3.head())

120. How Do You Get the Items Not Common to Both Series a and Series B?

>> #Input

>>import pandas as pd

>>ser1 = pd.Series([1, 2, 3, 4, 5])

>>ser2 = pd.Series([4, 5, 6, 7, 8])

>> #Solution

>>ser_u = pd.Series(np.union1d(ser1, ser2)) # union

>>ser_i = pd.Series(np.intersect1d(ser1, ser2)) # intersect

>ser_u[~ser_u.isin(ser_i)]

121. How Do You Keep Only the Top Two Most Frequent Values as It Is and Replace Everything Else as ‘other’ in a Series?

>> #Input

>>import pandas as pd

>>np.random.RandomState(100)

>>ser = pd.Series(np.random.randint(1, 5, [12]))

>> #Solution

>>print(“Top 2 Freq:”, ser.value_counts())

>>ser[~ser.isin(ser.value_counts().index[:2])] = ‘Other’

>ser

122. How Do You Find the Positions of Numbers That Are Multiples of Three from a Series?

>> #Input

>>import pandas as pd

>>ser = pd.Series(np.random.randint(1, 10, 7))

>>ser

>> #Solution

>>print(ser)

>np.argwhere(ser % 3==0)

123. How Do You Compute the Euclidean Distance Between Two Series?

The code is as shown:

>> #Input

>>p = pd.Series([1, 2, 3, 4, 5, 6, 7, 8, 9, 10])

>>q = pd.Series([10, 9, 8, 7, 6, 5, 4, 3, 2, 1])

>> #Solution

>>sum((p – q)**2)**.5

>> #Solution using func

>>np.linalg.norm(p-q)

You can see that the Euclidean distance can be calculated using two ways.

124. Which Python Library Is Built on Top of Matplotlib and Pandas to Ease Data Plotting?

Seaborn is a Python library built on top of matplotlib and pandas to ease data plotting. It is a data visualization library in Python that provides a high-level interface for drawing statistical informative graphs.

Did you know the answers to these Python interview questions? If not, here is what you can do.

125. What are the important features of Python?

• Python is a scripting language. Python, unlike other programming languages like C and its derivatives, does not require compilation prior to execution.
• Python is dynamically typed, which means you don’t have to specify the kinds of variables when declaring them or anything.
• Python is well suited to object-oriented programming since it supports class definition, composition, and inheritance.

126. What is PEP 8?

PEP denotes Python Enhancement Proposal. It’s a collection of guidelines for formatting Python code for maximum readability.

127. Explain Python namespace.

In Python, a namespace refers to the name that is assigned to each object.

128. What are decorators in Python?

Decorators are used for changing the appearance of a function without changing its structure. Decorators are typically defined prior to the function they are enhancing.

Decorators are typically defined prior to the function they are enhancing. To use a decorator, we must first specify its function. Then we write the function to which it is applied, simply placing the decorator function above the function to which it must be applied.

129. What is slicing and how to use in Python?

Slicing is a technique for gaining access to specific bits of sequences such as strings, tuples, and lists.

Slicing is a technique for gaining access to specific bits of sequences such as lists, tuples, and strings. The slicing syntax is [start:end:step]. This step can also be skipped. [start:end] returns all sequence items from the start (inclusive) to the end-1 element. It means the ith element from the end of the start or end element is negative i. The step represents the jump or the number of components that must be skipped.

130. How to combine dataframes in Pandas?

The following are the ways through which the data frames in Pandas can be combined:

• Concatenating them by vertically stacking the two dataframes.
• Concatenating them by horizontally stacking the two dataframes.
• Combining them on a common column. This is referred to as joining.

The concat() function is used to concatenate two dataframes. Its syntax is- pd.concat([dataframe1, dataframe2]).

Dataframes are joined together on a common column called a key. When we combine all the rows in data frame it is union and the join used is outer join. While, when we combine the common rows or intersection, the join used is the inner join. Its syntax is- pd.concat([dataframe1, dataframe2], axis=’axis’, join=’type_of_join)

• <li>Append():

This function is used for the horizontal stacking of data frames.

• concat(): This function is used for vertical stacking and best suites when the data frames to be combined possess the same column and similar fields.
• join(): This function is used to extract data from different data frames which have one or more columns common.

131. What are the key features of the Python 3.9.0.0 version?

• Zoneinfo and graphlib are two new modules.
• Improved modules such as asyncio and ast.
• Optimizations include improved idiom for assignment, signal handling, and Python built-ins.
• Removal of erroneous methods and functions.
• Instead of LL1, a new parser is based on PEG.
• Remove Prefixes and Suffixes with New String Methods.
• Generics with type hinting in standard collections.

132. Explain global variables and local variables in Python.

Local Variables:

A local variable is any variable declared within a function. This variable exists only in local space, not in global space.

Global Variables:

Global variables are variables declared outside of a function or in a global space. Any function in the program can access these variables.

133. How to install Python on Windows and set path variables?

• Download Python from https://www.python.org/downloads/
• Install it on your computer. Using your command prompt, look for the location where PYTHON is installed on your computer by typing cmd python.
• Then, in advanced system settings, create a new variable called PYTHON_NAME and paste the copied path into it.
• Search the path variable, choose its value and select ‘edit’.
• If the value doesn’t have a semicolon at the end, add one, and then type %PYTHON HOME%.

134. Is it necessary to indent in Python?

Indentation is required in Python. It designates a coding block. An indented block contains all of the code for loops, classes, functions, and so on. Typically, four space characters are used. Your code will not execute correctly if it is not indented, and it will also generate errors.

135. On Unix, how do you make a Python script executable?

Script file should start with #!/usr/bin/env python.

136. What are literals and the types of literals in Python?

For primitive data types, a literal in Python source code indicates a fixed value.

For primitive data types, a literal in Python source code indicates a fixed value. Following are the 5 types of literal in Python:

• String Literal: A string literal is formed by assigning some text to a variable that is contained in single or double-quotes. Assign the multiline text encased in triple quotes to produce multiline literals.
• Numeric Literal: They may contain numeric values that are floating-point values, integers, or complex numbers.
• Character Literal: It is made by putting a single character in double-quotes.
• Boolean Literal: True or False
• Literal Collections: There are four types of literals such as list collections, tuple literals, set literals, dictionary literals, and set literals.

137. How does continue, break, and pass work?

 

Continue	When a specified condition is met, the control is moved to the beginning of the loop, allowing some parts of the loop to be transferred.
Break	When a condition is met, the loop is terminated and control is passed to the next statement.
Pass	When you need a piece of code syntactically but don’t want to execute it, use this. This is a null operation.

 

138. In Python, are arguments provided by value or reference?

Pass by value: The actual item’s copy is passed. Changing the value of the object’s copy has no effect on the original object’s value.

Pass by reference: The actual object is passed as a reference. The value of the old object will change if the value of the new object is changed.

Arguments are passed by reference in Python.

def appendNumber(arr):

arr.append(4)

arr = [1, 2, 3]

print(arr)  #Output: => [1, 2, 3]

appendNumber(arr)

print(arr)  #Output: => [1, 2, 3, 4]

139. Explain join() and split() functions in Python.

The join() function can be used to combine a list of strings based on a delimiter into a single string.

The split() function can be used to split a string into a list of strings based on a delimiter.

string = “This is a string.”

string_list = string.split(‘ ‘) #delimiter is ‘space’ character or ‘ ‘

print(string_list) #output: [‘This’, ‘is’, ‘a’, ‘string.’]

print(‘ ‘.join(string_list)) #output: This is a string.

140. What are negative indexes and why are they used?

• The indexes from the end of the list, tuple, or string are called negative indexes.
• Arr[-1] denotes the array’s last element. Arr[]

 

141. In Python, how do you remark numerous lines?

Comments that involve multiple lines are known as multi-line comments. A # must prefix all lines that will be commented. You can also use a convenient shortcut to remark several lines. All you have to do is hold down the ctrl key and left-click anywhere you want a # character to appear, then input a # once. This will add a comment to every line where you put your cursor.

142. What is the purpose of ‘not’, ‘is’, and ‘in’ operators?

Special functions are known as operators. They take one or more input values and output a result.

not- returns the boolean value’s inverse

is- returns true when both operands are true

in- determines whether a certain element is present in a series

143. What are the functions help() and dir() used for in Python?

Both help() and dir() are available from the Python interpreter and are used to provide a condensed list of built-in functions.

dir() function: The defined symbols are displayed using the dir() function.

help() function: The help() function displays the documentation string and also allows you to access help for modules, keywords, attributes, and other items.

144. Why isn’t all the memory de-allocated when Python exits?

• When Python quits, some Python modules, especially those with circular references to other objects or objects referenced from global namespaces, are not necessarily freed or deallocated.
• Python would try to de-allocate/destroy all other objects on exit because it has its own efficient cleanup mechanism.
• It is difficult to de-allocate memory that has been reserved by the C library.

145. In Python, how do you utilize ternary operators?

The Ternary operator is the operator for displaying conditional statements. This is made of true or false values and a statement that must be evaluated.

146. Explain the split(), sub(), and subn() methods of the Python “re” module.

Python’s “re” module provides three ways for modifying strings. They are:

split (): a regex pattern is used to “separate” a string into a list

subn(): It works similarly to sub(), returning the new string as well as the number of replacements.

sub(): identifies all substrings that match the regex pattern and replaces them with a new string

147. What are negative indexes and why do we utilize them?

Python sequences are indexed, and they include both positive and negative values. Positive numbers are indexed with ‘0’ as the first index and ‘1’ as the second index, and so on.

The index for a negative number begins with ‘-1,’ which is the last index in the sequence, and ends with ‘-2,’ which is the penultimate index, and the sequence continues like a positive number. The negative index is used to eliminate all new-line spaces from the string and allow it to accept the last character S[:-1]. The negative index can also be used to represent the correct order of the string.

148. What are built-in types of Python?

Given below are the built-in types of Python:

• Built in functions
• Boolean
• String
• Complex numbers
• Floating point
• Integers

149. What are the benefits of NumPy arrays over (nested) Python lists?

• Lists in Python are useful general-purpose containers. They allow for (relatively) quick insertion, deletion, appending, and concatenation, and Python’s list comprehensions make them simple to create and operate.
• They have some limitations: they don’t enable “vectorized” operations like elementwise addition and multiplication, and because they can include objects of different types, Python must maintain type information for each element and execute type dispatching code while working on it.
• NumPy arrays are faster, and NumPy comes with a number of features, including histograms, algebra, linear, basic statistics, fast searching, convolutions, FFTs, and more.

 

150. What is the best way to add or remove values from a Python array?

The append(), extend(), and insert (i,x) procedures can be used to add elements to an array.y

The pop() and remove() methods can be used to remove elements from an array. The difference between these two functions is that one returns the removed value while the other does not.

151. Is there an object-oriented Programming (OOps) concept in Python?

Python is a computer language that focuses on objects. This indicates that by simply constructing an object model, every program can be solved in Python. Python, on the other hand, may be used as both a procedural and structured language.

152. Explain monkey patching in Python.

Monkey patches are solely used in Python to run-time dynamic updates to a class or module.

153. What is inheritance and different types of inheritance in Python?

Inheritance allows one class to gain all of another class’s members (for example, attributes and methods). Inheritance allows for code reuse, making it easier to develop and maintain applications.

The following are the various types of inheritance in Python:

• Single inheritance: The members of a single super class are acquired by a derived class.
• Multiple inheritance: More than one base class is inherited by a derived class.
• Muti-level inheritance: D1 is a derived class inherited from base1 while D2 is inherited from base2.
• Hierarchical Inheritance: You can inherit any number of child classes from a single base class.

 

154. Is multiple inheritance possible in Python?

A class can be inherited from multiple parent classes, which is known as multiple inheritance. In contrast to Java, Python allows multiple inheritance.

155. Explain polymorphism in Python.

The ability to take various forms is known as polymorphism. For example, if the parent class has a method named ABC, the child class can likewise have a method named ABC with its own parameters and variables. Python makes polymorphism possible.

156. What is encapsulation in Python?

Encapsulation refers to the joining of code and data. Encapsulation is demonstrated through a Python class.

157. In Python, how do you abstract data?

Only the necessary details are provided, while the implementation is hidden from view. Interfaces and abstract classes can be used to do this in Python.

158. Is access specifiers used in Python?

Access to an instance variable or function is not limited in Python. To imitate the behavior of protected and private access specifiers, Python introduces the idea of prefixing the name of the variable, function, or method with a single or double underscore.

159. How to create an empty class in Python?

A class that has no code defined within its block is called an empty class. The pass keyword can be used to generate it. You can, however, create objects of this class outside of the class. When used in Python, the PASS command has no effect.

160. What does an object() do?

It produces a featureless object that serves as the foundation for all classes. It also does not accept any parameters.

161. Write a Python program to generate a Star triangle.

 

1 2 3 4

def pyfunc(r): for x in range(r): print(‘ ‘*(r-x-1)+’*’*(2*x+1)) pyfunc(9)

Output:

*

***

*****

*******

*********

***********

*************

***************

*****************

162. Write a program to produce the Fibonacci series in Python.

 

1 2 3 4 5 6 7 8 9 10 11 12

# Enter number of terms needednbsp;#0,1,1,2,3,5…. a=int(input(“Enter the terms”)) f=0;#first element of series s=1#second element of series if a=0: print(“The requested series is”,f) else: print(f,s,end=” “) for x in range(2,a): print(next,end=” “) f=s s=next

Output: Enter the terms 5 0 1 1 2 3

163. Make a Python program that checks if a sequence is a Palindrome.

a=input(“enter sequence”)

b=a[::-1]

if a==b:

print(“palindrome”)

else:

print(“Not a Palindrome”)

Output: enter sequence 323 palindrome

164. Make a one-liner that counts how many capital letters are in a file. Even if the file is too large to fit in memory, your code should work.

 

1 2 3 4 5 6

with open(SOME_LARGE_FILE) as fh: count = 0 text = fh.read() for character in text: if character.isupper(): count += 1

Let us transform this into a single line

1	count sum(1 for line in fh for character in line if character.isupper())

165. Can you write a sorting algorithm with a numerical dataset?

 

1 2 3 4

list = [“1”, “4”, “0”, “6”, “9”] list = [int(i) for i in list] list.sort() print (list)

166. Check code given below, list the final value of A0, A1 …An.

 

1 2 3 4 5 6 7

A0 = dict(zip((‘a’,’b’,’c’,’d’,’e’),(1,2,3,4,5))) A1 = range(10)A2 = sorted([i for i in A1 if i in A0]) A3 = sorted([A0[s] for s in A0]) A4 = [i for i in A1 if i in A3] A5 = {i:i*i for i in A1} A6 = [[i,i*i] for i in A1] print(A0,A1,A2,A3,A4,A5,A6)

Here’s the answer:

A0 = {‘a’: 1, ‘c’: 3, ‘b’: 2, ‘e’: 5, ‘d’: 4} # the order may vary

A1 = range(0, 10)

A2 = []

A3 = [1, 2, 3, 4, 5]

A4 = [1, 2, 3, 4, 5]

A5 = {0: 0, 1: 1, 2: 4, 3: 9, 4: 16, 5: 25, 6: 36, 7: 49, 8: 64, 9: 81}

A6 = [[0, 0], [1, 1], [2, 4], [3, 9], [4, 16], [5, 25], [6, 36], [7, 49], [8, 64], [9, 81]]

167. In NumPy, how will you read CSV data into an array?

This may be accomplished by utilizing the genfromtxt() method with a comma as the delimiter.

168. What is GIL?

The term GIL stands for Global Interpreter Lock. This is a mutex that helps thread synchronization by preventing deadlocks by limiting access to Python objects. GIL assists with multitasking (and not parallel computing).

169. What is PIP?

PIP denotes Python Installer Package. It is used to install various Python modules. It’s a command-line utility that creates a unified interface for installing various Python modules. It searches the internet for the package and installs it into the working directory without requiring any user intervention.

170. Write a program that checks if all of the numbers in a sequence are unique.

 

def check_distinct(data_list):

if len(data_list) == len(set(data_list)):

return True

else:

return False;

print(check_distinct([1,6,5,8]))     #Prints True

print(check_distinct([2,2,5,5,7,8])) #Prints False

171. What is an operator in Python?

An operator is a symbol that is applied to a set of values to produce a result. An operator manipulates operands. Numeric literals or variables that hold values are known as operands. Unary, binary, and ternary operators are all possible. The unary operator, which requires only one operand, the binary operator, which requires two operands, and the ternary operator, which requires three operands.

172. What are the various types of operators in Python?

• Bitwise operators
• Identity operators
• Membership operators
• Logical operators
• Assignment operators
• Relational operators
• Arithmetic operators

173. How to write a Unicode string in Python?

The old Unicode type has been replaced with the “str” type in Python 3, and the string is now considered Unicode by default. Using the art.title.encode(“utf-8”) function, we can create a Unicode string.

174. How to send an email in Python language?

Python includes the smtplib and email libraries for sending emails. Import these modules into the newly generated mail script and send mail to users who have been authenticated.

175. Create a program to add two integers >0 without using the plus operator.

def add_nums(num1, num2):

while num2 != 0:

data = num1 & num2

num1 = num1 ^ num2

num2 = data << 1

return num1

print(add_nums(2, 10))

176. Create a program to convert dates from yyyy-mm-dd to dd-mm-yyyyy.

We can use this module to convert dates:

import re

def transform_date_format(date):

return re.sub(r'(\d{4})-(\d{1,2})-(\d{1,2})’, ‘\\3-\\2-\\1’, date)

date_input = “2021-08-01”

print(transform_date_format(date_input))

The datetime module can also be used, as demonstrated below:

from datetime import datetime

new_date = datetime.strptime(“2021-08-01”, “%Y-%m-%d”).strftime(“%d:%m:%Y”)

print(new_data)

177. Create a program that combines two dictionaries. If you locate the same keys during combining, you can sum the values of these similar keys. Create a new dictionary.

from collections import Counter

d1 = {‘key1’: 50, ‘key2’: 100, ‘key3’:200}

d2 = {‘key1’: 200, ‘key2’: 100, ‘key4’:300}

new_dict = Counter(d1) + Counter(d2)

print(new_dict)

178. What kind of joins are offered by Pandas?

There are four joins in Pandas: left, inner, right, and outer.

179. How are dataframes in Pandas merged?

The type and fields of the dataframes being merged determine how they are merged. If the data has identical fields, it is combined along axis 0, otherwise, it is merged along axis 1.

180. What are generators and decorators in python?

Functions which return an iterable set of items are known as generators.

Generator functions act just like regular functions with just one difference they use the Python yield keyword instead of return.

A generator function is a function that returns an iterator. A generator expression is an expression that returns an iterator.

Generator objects are used either by calling the next method on the generator object or using the generator object in a “for in” loop.

 

A decorator in Python is any callable Python object that is used to modify a function or a class. It takes in a function, adds some functionality, and returns it.

Decorators are a very powerful and useful tool in Python since it allows programmers to modify/control the behaviour of a function or class.

Decorators are usually called before the definition of a function you want to decorate. There are two different kinds of decorators in Python:

 

Function decorators

Class decorators

 

181. Difference b/w numpy arrays and list

Arrays:

Arrays need to be imported using an array or numpy.

The collection of multiple items of the same data type in an array is another essential element.

Arrays need to have the same size as the array for nesting.

To print the array elements, a specific loop must be created.

An array consumes less memory in comparison to a list.

The array is not flexible because it makes it difficult to modify data.

We can apply direct arithmetic operations in an array.

Supports broadcasting, element-wise operations

Homogeneous (all elements of the same type)

Created using NumPy.array

 

 

Lists:

Lists are in-built data structures.

Lists collect items that typically have components of various data types.

In list variable size nesting is possible.

Without explicitly creating a loop, the List can be printed.

The List consumes more memory.

The List is ideal in terms of flexibility since it makes data modification simple.

In lists, it is not possible to apply arithmetic operations directly.

Does not support broadcasting efficiently

Heterogeneous (elements can have any type)

Created using square brackets [ ]

Machine Learning

1. What Are the Different Types of Machine Learning?

There are three types of machine learning:

Supervised Learning

In supervised machine learning, a model makes predictions or decisions based on past or labeled data. Labeled data refers to sets of data that are given tags or labels, and thus made more meaningful.

Unsupervised Learning

In unsupervised learning, we don’t have labeled data. A model can identify patterns, anomalies, and relationships in the input data.

Reinforcement Learning

Using reinforcement learning, the model can learn based on the rewards it received for its previous action.

Consider an environment where an agent is working. The agent is given a target to achieve. Every time the agent takes some action toward the target, it is given positive feedback. And, if the action taken is going away from the goal, the agent is given negative feedback.

 

2. What is Overfitting, and How Can You Avoid It?

The Overfitting is a situation that occurs when a model learns the training set too well, taking up random fluctuations in the training data as concepts. These impact the model’s ability to generalize and don’t apply to new data.

When a model is given the training data, it shows 100 percent accuracy—technically a slight loss. But, when we use the test data, there may be an error and low efficiency. This condition is known as overfitting.

There are multiple ways of avoiding overfitting, such as:

• Regularization. It involves a cost term for the features involved with the objective function
• Making a simple model. With lesser variables and parameters, the variance can be reduced
• Cross-validation methods like k-folds can also be used
• If some model parameters are likely to cause overfitting, techniques for regularization like LASSO can be used that penalize these parameters

3. What is ‘training Set’ and ‘test Set’ in a Machine Learning Model? How Much Data Will You Allocate for Your Training, Validation, and Test Sets?

There is a three-step process followed to create a model:

1. Train the model
2. Test the model
3. Deploy the model

Training Set	Test Set
The training set is examples given to the model to analyze and learn 70% of the total data is typically taken as the training dataset This is labeled data used to train the model	The test set is used to test the accuracy of the hypothesis generated by the model Remaining 30% is taken as testing dataset We test without labeled data and then verify results with labels

Consider a case where you have labeled data for 1,000 records. One way to train the model is to expose all 1,000 records during the training process. Then you take a small set of the same data to test the model, which would give good results in this case.

But, this is not an accurate way of testing. So, we set aside a portion of that data called the ‘test set’ before starting the training process. The remaining data is called the ‘training set’ that we use for training the model. The training set passes through the model multiple times until the accuracy is high, and errors are minimized.

Now, we pass the test data to check if the model can accurately predict the values and determine if training is effective. If you get errors, you either need to change your model or retrain it with more data.

Regarding the question of how to split the data into a training set and test set, there is no fixed rule, and the ratio can vary based on individual preferences.

4. How Do You Handle Missing or Corrupted Data in a Dataset?

One of the easiest ways to handle missing or corrupted data is to drop those rows or columns or replace them entirely with some other value.

There are two useful methods in Pandas:

• IsNull() and dropna() will help to find the columns/rows with missing data and drop them
• Fillna() will replace the wrong values with a placeholder value

5. How Can You Choose a Classifier Based on a Training Set Data Size?

When the training set is small, a model that has a right bias and low variance seems to work better because they are less likely to overfit.

For example, Naive Bayes works best when the training set is large. Models with low bias and high variance tend to perform better as they work fine with complex relationships.

6. Explain the Confusion Matrix with Respect to Machine Learning Algorithms.

A confusion matrix (or error matrix) is a specific table that is used to measure the performance of an algorithm. It is mostly used in supervised learning; in unsupervised learning, it’s called the matching matrix.

The confusion matrix has two parameters:

• Actual
• Predicted

It also has identical sets of features in both of these dimensions.

Consider a confusion matrix (binary matrix) shown below:

Here,

For actual values:

Total Yes = 12+1 = 13

Total No = 3+9 = 12

Similarly, for predicted values:

Total Yes = 12+3 = 15

Total No = 1+9 = 10

For a model to be accurate, the values across the diagonals should be high. The total sum of all the values in the matrix equals the total observations in the test data set.

For the above matrix, total observations = 12+3+1+9 = 25

Now, accuracy = sum of the values across the diagonal/total dataset

= (12+9) / 25

= 21 / 25

= 84%

7. What Is a False Positive and False Negative and How Are They Significant?

False positives are those cases that wrongly get classified as True but are False.

False negatives are those cases that wrongly get classified as False but are True.

In the term ‘False Positive,’ the word ‘Positive’ refers to the ‘Yes’ row of the predicted value in the confusion matrix. The complete term indicates that the system has predicted it as a positive, but the actual value is negative.

So, looking at the confusion matrix, we get:

False-positive = 3

True positive = 12

Similarly, in the term ‘False Negative,’ the word ‘Negative’ refers to the ‘No’ row of the predicted value in the confusion matrix. And the complete term indicates that the system has predicted it as negative, but the actual value is positive.

So, looking at the confusion matrix, we get:

False Negative = 1

True Negative = 9

8. What Are the Three Stages of Building a Model in Machine Learning?

The three stages of building a machine learning model are:

• Model Building

  Choose a suitable algorithm for the model and train it according to the requirement

• Model Testing

  Check the accuracy of the model through the test data

• Applying the Model

  Make the required changes after testing and use the final model for real-time projects

Here, it’s important to remember that once in a while, the model needs to be checked to make sure it’s working correctly. It should be modified to make sure that it is up-to-date.

9. What is Deep Learning?

The Deep learning is a subset of machine learning that involves systems that think and learn like humans using artificial neural networks. The term ‘deep’ comes from the fact that you can have several layers of neural networks.

One of the primary differences between machine learning and deep learning is that feature engineering is done manually in machine learning. In the case of deep learning, the model consisting of neural networks will automatically determine which features to use (and which not to use).

This is a commonly asked question asked in both Machine Learning Interviews as well as Deep Learning Interview Questions

10. What Are the Differences Between Machine Learning and Deep Learning?

Learn more: Difference Between AI,ML and Deep Learning

Machine Learning	Deep Learning
Enables machines to take decisions on their own, based on past data It needs only a small amount of data for training Works well on the low-end system, so you don’t need large machines Most features need to be identified in advance and manually coded The problem is divided into two parts and solved individually and then combined	Enables machines to take decisions with the help of artificial neural networks It needs a large amount of training data Needs high-end machines because it requires a lot of computing power The machine learns the features from the data it is provided The problem is solved in an end-to-end manner

11. What Are the Applications of Supervised Machine Learning in Modern Businesses?

Applications of supervised machine learning include:

• Email Spam Detection

  Here we train the model using historical data that consists of emails categorized as spam or not spam. This labeled information is fed as input to the model.

• Healthcare Diagnosis

  By providing images regarding a disease, a model can be trained to detect if a person is suffering from the disease or not.

• Sentiment Analysis

  This refers to the process of using algorithms to mine documents and determine whether they’re positive, neutral, or negative in sentiment.

• Fraud Detection

  By training the model to identify suspicious patterns, we can detect instances of possible fraud.

12. What is Semi-supervised Machine Learning?

Supervised learning uses data that is completely labeled, whereas unsupervised learning uses no training data.

In the case of semi-supervised learning, the training data contains a small amount of labeled data and a large amount of unlabeled data.

13. What Are Unsupervised Machine Learning Techniques? 

There are two techniques used in unsupervised learning: clustering and association.

Clustering

Clustering problems involve data to be divided into subsets. These subsets, also called clusters, contain data that are similar to each other. Different clusters reveal different details about the objects, unlike classification or regression.

Association

In an association problem, we identify patterns of associations between different variables or items.

For example, an e-commerce website can suggest other items for you to buy, based on the prior purchases that you have made, spending habits, items in your wishlist, other customers’ purchase habits, and so on.

14. What is the Difference Between Supervised and Unsupervised Machine Learning?

• Supervised learning – This model learns from the labeled data and makes a future prediction as output
• Unsupervised learning – This model uses unlabeled input data and allows the algorithm to act on that information without guidance.

 

15. What is the Difference Between Inductive Machine Learning and Deductive Machine Learning? 

Inductive Learning	Deductive Learning
It observes instances based on defined principles to draw a conclusion Example: Explaining to a child to keep away from the fire by showing a video where fire causes damage	It concludes experiences Example: Allow the child to play with fire. If he or she gets burned, they will learn that it is dangerous and will refrain from making the same mistake again

16. Compare K-means and KNN Algorithms.

K-means	KNN
K-Means is unsupervised K-Means is a clustering algorithm The points in each cluster are similar to each other, and each cluster is different from its neighboring clusters	KNN is supervised in nature KNN is a classification algorithm It classifies an unlabeled observation based on its K (can be any number) surrounding neighbors

17. What Is ‘naive’ in the Naive Bayes Classifier?

The classifier is called ‘naive’ because it makes assumptions that may or may not turn out to be correct.

The algorithm assumes that the presence of one feature of a class is not related to the presence of any other feature (absolute independence of features), given the class variable.

For instance, a fruit may be considered to be a cherry if it is red in color and round in shape, regardless of other features. This assumption may or may not be right (as an apple also matches the description).

18. Explain How a System Can Play a Game of Chess Using Reinforcement Learning.

Reinforcement learning has an environment and an agent. The agent performs some actions to achieve a specific goal. Every time the agent performs a task that is taking it towards the goal, it is rewarded. And, every time it takes a step that goes against that goal or in the reverse direction, it is penalized.

Earlier, chess programs had to determine the best moves after much research on numerous factors. Building a machine designed to play such games would require many rules to be specified.

With reinforced learning, we don’t have to deal with this problem as the learning agent learns by playing the game. It will make a move (decision), check if it’s the right move (feedback), and keep the outcomes in memory for the next step it takes (learning). There is a reward for every correct decision the system takes and punishment for the wrong one.

19. How Will You Know Which Machine Learning Algorithm to Choose for Your Classification Problem?

While there is no fixed rule to choose an algorithm for a classification problem, you can follow these guidelines:

• If accuracy is a concern, test different algorithms and cross-validate them
• If the training dataset is small, use models that have low variance and high bias
• If the training dataset is large, use models that have high variance and little bias

20. How is Amazon Able to Recommend Other Things to Buy? How Does the Recommendation Engine Work?

Once a user buys something from Amazon, Amazon stores that purchase data for future reference and finds products that are most likely also to be bought, it is possible because of the Association algorithm, which can identify patterns in a given dataset.

21. When Will You Use Classification over Regression?

Classification is used when your target is categorical, while regression is used when your target variable is continuous. Both classification and regression belong to the category of supervised machine learning algorithms.

Examples of classification problems include:

• Predicting yes or no
• Estimating gender
• Breed of an animal
• Type of color

Examples of regression problems include:

• Estimating sales and price of a product
• Predicting the score of a team
• Predicting the amount of rainfall

22. How Do You Design an Email Spam Filter?

Building a spam filter involves the following process:

• The email spam filter will be fed with thousands of emails
• Each of these emails already has a label: ‘spam’ or ‘not spam.’
• The supervised machine learning algorithm will then determine which type of emails are being marked as spam based on spam words like the lottery, free offer, no money, full refund, etc.
• The next time an email is about to hit your inbox, the spam filter will use statistical analysis and algorithms like Decision Trees and SVM to determine how likely the email is spam
• If the likelihood is high, it will label it as spam, and the email won’t hit your inbox
• Based on the accuracy of each model, we will use the algorithm with the highest accuracy after testing all the models

23. What is a Random Forest?

A ‘random forest’ is a supervised machine learning algorithm that is generally used for classification problems. It operates by constructing multiple decision trees during the training phase. The random forest chooses the decision of the majority of the trees as the final decision.

24. Considering a Long List of Machine Learning Algorithms, given a Data Set, How Do You Decide Which One to Use?

There is no master algorithm for all situations. Choosing an algorithm depends on the following questions:

• How much data do you have, and is it continuous or categorical?
• Is the problem related to classification, association, clustering, or regression?
• Predefined variables (labeled), unlabeled, or mix?
• What is the goal?

Based on the above questions, the following algorithms can be used:

25. What is Bias and Variance in a Machine Learning Model?

Bias

Bias in a machine learning model occurs when the predicted values are further from the actual values. Low bias indicates a model where the prediction values are very close to the actual ones.

Underfitting: High bias can cause an algorithm to miss the relevant relations between features and target outputs.

Variance

Variance refers to the amount the target model will change when trained with different training data. For a good model, the variance should be minimized.

Overfitting: High variance can cause an algorithm to model the random noise in the training data rather than the intended outputs.

26. What is the Trade-off Between Bias and Variance?

The bias-variance decomposition essentially decomposes the learning error from any algorithm by adding the bias, variance, and a bit of irreducible error due to noise in the underlying dataset.

Necessarily, if you make the model more complex and add more variables, you’ll lose bias but gain variance. To get the optimally-reduced amount of error, you’ll have to trade off bias and variance. Neither high bias nor high variance is desired.

High bias and low variance algorithms train models that are consistent, but inaccurate on average.

High variance and low bias algorithms train models that are accurate but inconsistent.

27. Define Precision and Recall.

Precision

Precision is the ratio of several events you can correctly recall to the total number of events you recall (mix of correct and wrong recalls).

Precision = (True Positive) / (True Positive + False Positive)

Recall

A recall is the ratio of the number of events you can recall the number of total events.

Recall = (True Positive) / (True Positive + False Negative)

28. What is a Decision Tree Classification?

A decision tree builds classification (or regression) models as a tree structure, with datasets broken up into ever-smaller subsets while developing the decision tree, literally in a tree-like way with branches and nodes. Decision trees can handle both categorical and numerical data.

29. What is Pruning in Decision Trees, and How Is It Done?

Pruning is a technique in machine learning that reduces the size of decision trees. It reduces the complexity of the final classifier, and hence improves predictive accuracy by the reduction of overfitting.

Pruning can occur in:

• Top-down fashion. It will traverse nodes and trim subtrees starting at the root
• Bottom-up fashion. It will begin at the leaf nodes

There is a popular pruning algorithm called reduced error pruning, in which:

• Starting at the leaves, each node is replaced with its most popular class
• If the prediction accuracy is not affected, the change is kept
• There is an advantage of simplicity and speed

30. Briefly Explain Logistic Regression.

Logistic regression is a classification algorithm used to predict a binary outcome for a given set of independent variables.

The output of logistic regression is either a 0 or 1 with a threshold value of generally 0.5. Any value above 0.5 is considered as 1, and any point below 0.5 is considered as 0.

31. Explain the K Nearest Neighbor Algorithm. 

K nearest neighbor algorithm is a classification algorithm that works in a way that a new data point is assigned to a neighboring group to which it is most similar.

In K nearest neighbors, K can be an integer greater than 1. So, for every new data point, we want to classify, we compute to which neighboring group it is closest.

Let us classify an object using the following example. Consider there are three clusters:

• Football
• Basketball
• Tennis ball

Let the new data point to be classified is a black ball. We use KNN to classify it. Assume K = 5 (initially).

Next, we find the K (five) nearest data points, as shown.

Observe that all five selected points do not belong to the same cluster. There are three tennis balls and one each of basketball and football.

When multiple classes are involved, we prefer the majority. Here the majority is with the tennis ball, so the new data point is assigned to this cluster.

32. What is a Recommendation System?

Anyone who has used Spotify or shopped at Amazon will recognize a recommendation system: It’s an information filtering system that predicts what a user might want to hear or see based on choice patterns provided by the user.

33. What is Kernel SVM?

Kernel SVM is the abbreviated version of the kernel support vector machine. Kernel methods are a class of algorithms for pattern analysis, and the most common one is the kernel SVM.

34. What Are Some Methods of Reducing Dimensionality?

You can reduce dimensionality by combining features with feature engineering, removing collinear features, or using algorithmic dimensionality reduction.

Now that you have gone through these machine learning interview questions, you must have got an idea of your strengths and weaknesses in this domain.

35. What is Principal Component Analysis?

Principal Component Analysis or PCA is a multivariate statistical technique that is used for analyzing quantitative data. The objective of PCA is to reduce higher dimensional data to lower dimensions, remove noise, and extract crucial information such as features and attributes from large amounts of data.

36. What do you understand by the F1 score?

The F1 score is a metric that combines both Precision and Recall. It is also the weighted average of precision and recall.

The F1 score can be calculated using the below formula:

F1 = 2 * (P * R) / (P + R)

The F1 score is one when both Precision and Recall scores are one.

37. What do you understand by Type I vs Type II error?

Type I Error: Type I error occurs when the null hypothesis is true and we reject it.

Type II Error: Type II error occurs when the null hypothesis is false and we accept it.

38. Explain Correlation and Covariance?

Correlation: Correlation tells us how strongly two random variables are related to each other. It takes values between -1 to +1.

Formula to calculate Correlation:

Covariance: Covariance tells us the direction of the linear relationship between two random variables. It can take any value between – ∞ and + ∞.

Formula to calculate Covariance:

39. What are Support Vectors in SVM?

Support Vectors are data points that are nearest to the hyperplane. It influences the position and orientation of the hyperplane. Removing the support vectors will alter the position of the hyperplane. The support vectors help us build our support vector machine model.

40. What is Ensemble learning?

Ensemble learning is a combination of the results obtained from multiple machine learning models to increase the accuracy for improved decision-making.

Example: A Random Forest with 100 trees can provide much better results than using just one decision tree.

41. What is Cross-Validation?

Cross-Validation in Machine Learning is a statistical resampling technique that uses different parts of the dataset to train and test a machine learning algorithm on different iterations. The aim of cross-validation is to test the model’s ability to predict a new set of data that was not used to train the model. Cross-validation avoids the overfitting of data.

K-Fold Cross Validation is the most popular resampling technique that divides the whole dataset into K sets of equal sizes.

42. What are the different methods to split a tree in a decision tree algorithm?

Variance: Splitting the nodes of a decision tree using the variance is done when the target variable is continuous.

Information Gain: Splitting the nodes of a decision tree using Information Gain is preferred when the target variable is categorical.

Gini Impurity: Splitting the nodes of a decision tree using Gini Impurity is followed when the target variable is categorical.

43. How does the Support Vector Machine algorithm handle self-learning? 

The SVM algorithm has a learning rate and expansion rate which takes care of self-learning. The learning rate compensates or penalizes the hyperplanes for making all the incorrect moves while the expansion rate handles finding the maximum separation area between different classes.

44. What are the assumptions you need to take before starting with linear regression?

There are primarily 5 assumptions for a Linear Regression model:

• Multivariate normality
• No auto-correlation
• Homoscedasticity
• Linear relationship
• No or little multicollinearity

45. What is the difference between Lasso and Ridge regression?

Lasso(also known as L1) and Ridge(also known as L2) regression are two popular regularization techniques that are used to avoid overfitting of data. These methods are used to penalize the coefficients to find the optimum solution and reduce complexity. The Lasso regression works by penalizing the sum of the absolute values of the coefficients. In Ridge or L2 regression, the penalty function is determined by the sum of the squares of the coefficients.

 

46. How machine learning is different from general programming?

In general programming, we have the data and the logic by using these two we create the answers. But in machine learning, we have the data and the answers and we let the machine learn the logic from them so, that the same logic can be used to answer the questions which will be faced in the future.

Also, there are times when writing logic in codes is not possible so, at those times machine learning becomes a saviour and learns the logic itself.

47. What are some real-life applications of clustering algorithms?

The clustering technique can be used in multiple domains of data science like image classification, customer segmentation, and recommendation engine. One of the most common use is in market research and customer segmentation which is then utilized to target a particular market group to expand the businesses and profitable outcomes.

48. How to choose an optimal number of clusters?

By using the Elbow method we decide an optimal number of clusters that our clustering algorithm must try to form. The main principle behind this method is that if we will increase the number of clusters the error value will decrease.

But after an optimal number of features, the decrease in the error value is insignificant so, at the point after which this starts to happen, we choose that point as the optimal number of clusters that the algorithm will try to form.

The optimal number of clusters from the above figure is 3.

49. What is feature engineering? How does it affect the model’s performance? 

Feature engineering refers to developing some new features by using existing features. Sometimes there is a very subtle mathematical relation between some features which if explored properly then the new features can be developed using those mathematical operations.

Also, there are times when multiple pieces of information are clubbed and provided as a single data column. At those times developing new features and using them help us to gain deeper insights into the data as well as if the features derived are significant enough helps to improve the model’s performance a lot.

A hypothesis is a term that is generally used in the Supervised machine learning domain. As we have independent features and target variables and we try to find an approximate function mapping from the feature space to the target variable that approximation of mapping is known as a hypothesis.

51. How do measure the effectiveness of the clusters?

There are metrics like Inertia or Sum of Squared Errors (SSE), Silhouette Score, l1, and l2 scores. Out of all of these metrics, the Inertia or Sum of Squared Errors (SSE) and Silhouette score is a common metrics for measuring the effectiveness of the clusters.

Although this method is quite expensive in terms of computation cost. The score is high if the clusters formed are dense and well separated.

52. Why do we take smaller values of the learning rate?

Smaller values of learning rate help the training process to converge more slowly and gradually toward the global optimum instead of fluctuating around it. This is because a smaller learning rate results in smaller updates to the model weights at each iteration, which can help to ensure that the updates are more precise and stable.
If the learning rate is too large, the model weights can update too quickly, which can cause the training process to overshoot the global optimum and miss it entirely.

So, to avoid this oscillation of the error value and achieve the best weights for the model this is necessary to use smaller values of the learning rate.

53. What is Overfitting in Machine Learning and how can it be avoided?

Overfitting happens when the model learns patterns as well as the noises present in the data this leads to high performance on the training data but very low performance for data that the model has not seen earlier. To avoid overfitting there are multiple methods that we can use:

• Early stopping of the model’s training in case of validation training stops increasing but the training keeps going on.
• Using regularization methods like L1 or L2 regularization which is used to penalize the model’s weights to avoid overfitting.

54. Why we cannot use linear regression for a classification task?

The main reason why we cannot use linear regression for a classification task is that the output of linear regression is continuous and unbounded, while classification requires discrete and bounded output values.

If we use linear regression for the classification task the error function graph will not be convex. A convex graph has only one minimum which is also known as the global minima but in the case of the non-convex graph, there are chances of our model getting stuck at some local minima which may not be the global minima. To avoid this situation of getting stuck at the local minima we do not use the linear regression algorithm for a classification task.

55. What is the purpose of splitting a given dataset into training and validation data?

The main purpose is to keep some data left over on which the model has not been trained so, that we can evaluate the performance of our machine learning model after training. Also, sometimes we use the validation dataset to choose among the multiple state-of-the-art machine learning models. Like we first train some models let’s say LogisticRegression, XGBoost, or any other than test their performance using validation data and choose the model which has less difference between the validation and the training accuracy.

56. Why do we perform normalization?

To achieve stable and fast training of the model we use normalization techniques to bring all the features to a certain scale or range of values. If we do not perform normalization then there are chances that the gradient will not converge to the global or local minima and end up oscillating back and forth. Read more about it here.

57. What is the difference between precision and recall?

Precision is simply the ratio between the true positives(TP) and all the positive examples (TP+FP) predicted by the model. In other words, precision measures how many of the predicted positive examples are actually true positives. It is a measure of the model’s ability to avoid false positives and make accurate positive predictions.

But in the case of a recall, we calculate the ratio of true positives (TP) and the total number of examples (TP+FN) that actually fall in the positive class. recall measures how many of the actual positive examples are correctly identified by the model. It is a measure of the model’s ability to avoid false negatives and identify all positive examples correctly.

58. What is the difference between upsampling and downsampling?

In the upsampling method, we increase the number of samples in the minority class by randomly selecting some points from the minority class and adding them to the dataset repeat this process till the dataset gets balanced for each class. But here is a disadvantage the training accuracy becomes high as in each epoch model trained more than once in each epoch but the same high accuracy is not observed in the validation accuracy.

In the case of downsampling, we decrease the number of samples in the majority class by selecting some random number of points that are equal to the number of data points in the minority class so that the distribution becomes balanced. In this case, we have to suffer from data loss which may lead to the loss of some critical information as well.

59. What is data leakage and how can we identify it?

If there is a high correlation between the target variable and the input features then this situation is referred to as data leakage. This is because when we train our model with that highly correlated feature then the model gets most of the target variable’s information in the training process only and it has to do very little to achieve high accuracy. In this situation, the model gives pretty decent performance both on the training as well as the validation data but as we use that model to make actual predictions then the model’s performance is not up to the mark. This is how we can identify data leakage.

60. Explain the classification report and the metrics it includes.

Classification reports are evaluated using classification metrics that have precision, recall, and f1-score on a per-class basis.

• Precision can be defined as the ability of a classifier not to label an instance positive that is actually negative.
• Recall is the ability of a classifier to find all positive values. For each class, it is defined as the ratio of true positives to the sum of true positives and false negatives.
• F1-score is a harmonic mean of precision and recall.
• Support is the number of samples used for each class.
• The overall accuracy score of the model is also there to get a high-level review of the performance. It is the ratio between the total number of correct predictions and the total number of datasets.
• Macro avg is nothing but the average of the metric(precision, recall, f1-score) values for each class.
• The weighted average is calculated by providing a higher preference to that class that was present in the higher number in the datasets.

61. What are some of the hyperparameters of the random forest regressor which help to avoid overfitting?

The most important hyper-parameters of a Random Forest are:

• max_depth – Sometimes the larger depth of the tree can create overfitting. To overcome it, the depth should be limited.
• n-estimator – It is the number of decision trees we want in our forest.
• min_sample_split – It is the minimum number of samples an internal node must hold in order to split into further nodes.
• max_leaf_nodes – It helps the model to control the splitting of the nodes and in turn, the depth of the model is also restricted.

62. What is the bias-variance tradeoff?

First, let’s understand what is bias and variance:

• Bias refers to the difference between the actual values and the predicted values by the model. Low bias means the model has learned the pattern in the data and high bias means the model is unable to learn the patterns present in the data i.e the underfitting.
• Variance refers to the change in accuracy of the model’s prediction on which the model has not been trained. Low variance is a good case but high variance means that the performance of the training data and the validation data vary a lot.

If the bias is too low but the variance is too high then that case is known as overfitting. So, finding a balance between these two situations is known as the bias-variance trade-off.

No there is no such necessary condition that the data must be split into 80:20 ratio. The main purpose of the splitting is to have some data which the model has not seen previously so, that we can evaluate the performance of the model.

If the dataset contains let’s say 50,000 rows of data then only 1000 or maybe 2000 rows of data is enough to evaluate the model’s performance.

64. What is Principal Component Analysis?

PCA(Principal Component Analysis) is an unsupervised machine learning dimensionality reduction technique in which we trade off some information or patterns of the data at the cost of reducing its size significantly. In this algorithm, we try to preserve the variance of the original dataset up to a great extent let’s say 95%. For very high dimensional data sometimes even at the loss of 1% of the variance, we can reduce the data size significantly.

By using this algorithm we can perform image compression, visualize high-dimensional data as well as make data visualization easy.

65. What is one-shot learning?

One-shot learning is a concept in machine learning where the model is trained to recognize the patterns in datasets from a single example instead of training on large datasets. This is useful when we haven’t large datasets. It is applied to find the similarity and dissimilarities between the two images.

66. What is the difference between Manhattan Distance and Euclidean distance?

Both Manhattan Distance and Euclidean distance are two distance measurement techniques.

Manhattan Distance (MD) is calculated as the sum of absolute differences between the coordinates of two points along each dimension.

Euclidean Distance (ED) is calculated as the square root of the sum of squared differences between the coordinates of two points along each dimension.

Generally, these two metrics are used to evaluate the effectiveness of the clusters formed by a clustering algorithm.

67. What is the difference between covariance and correlation?

As the name suggests, Covariance provides us with a measure of the extent to which two variables differ from each other. But on the other hand, correlation gives us the measure of the extent to which the two variables are related to each other. Covariance can take on any value while correlation is always between -1 and 1. These measures are used during the exploratory data analysis to gain insights from the data.

68. What is the difference between one hot encoding and ordinal encoding?

One Hot encoding and ordinal encoding both are different methods to convert categorical features to numeric ones the difference is in the way they are implemented. In one hot encoding, we create a separate column for each category and add 0 or 1 as per the value corresponding to that row. Contrary to one hot encoding, In ordinal encoding, we replace the categories with numbers from 0 to n-1 based on the order or rank where n is the number of unique categories present in the dataset. The main difference between one-hot encoding and ordinal encoding is that one-hot encoding results in a binary matrix representation of the data in the form of 0 and 1, it is used when there is no order or ranking between the dataset whereas ordinal encoding represents categories as ordinal values.

69. How to identify whether the model has overfitted the training data or not?

This is the step where the splitting of the data into training and validation data proves to be a boon. If the model’s performance on the training data is very high as compared to the performance on the validation data then we can say that the model has overfitted the training data by learning the patterns as well as the noise present in the dataset.

70. How can you conclude about the model’s performance using the confusion matrix?

confusion matrix summarizes the performance of a classification model. In a confusion matrix, we get four types of output (in case of a binary classification problem) which are TP, TN, FP, and FN. As we know that there are two diagonals possible in a square, and one of these two diagonals represents the numbers for which our model’s prediction and the true labels are the same. Our target is also to maximize the values along these diagonals. From the confusion matrix, we can calculate various evaluation metrics like accuracy, precision, recall, F1 score, etc.

71. What is the use of the violin plot?

The name violin plot has been derived from the shape of the graph which matches the violin. This graph is an extension of the Kernel Density Plot along with the properties of the boxplot. All the statistical measures shown by a boxplot are also shown by the violin plot but along with this, The width of the violin represents the density of the variable in the different regions of values. This visualization tool is generally used in the exploratory data analysis step to check the distribution of the continuous data variables.

With this, we have covered some of the most important Machine Learning concepts which are generally asked by the interviewers to test the technical understanding of a candidate also, we would like to wish you all the best for your next interview.

72. What are the five statistical measures represented in a boxplot?

• Left Whisker – This statistical measure is calculated by subtracting 1.5 times IQR(Inter Quartile Range) from Q1.
  • IQR = Q3-Q1
  • Left Whisker = Q1-1.5*IQR
• Q1 – This is also known as the 25 percentile.
• Q2 – This is the median of the data or 50 percentile.
• Q3 – This is also known as 75 percentile
• Right Whisker – This statistical measure is calculated by adding 1.5 times of IQR(Inter Quartile Range) in Q3.
  • Right Whisker = Q3 + 1.5*IQR

73. What is the difference between stochastic gradient descent (SGD) and gradient descent (GD)?

In the gradient descent algorithm train our model on the whole dataset at once. But in Stochastic Gradient Descent, the model is trained by using a mini-batch of training data at once. If we are using SGD then one cannot expect the training error to go down smoothly. The training error oscillates but after some training steps, we can say that the training error has gone down. Also, the minima achieved by using GD may vary from that achieved using the SGD. It is observed that the minima achieved by using SGD are close to GD but not the same.

74. What is the Central Limit theorem?

This theorem is related to sampling statistics and its distribution. As per this theorem the sampling distribution of the sample means tends to towards a normal distribution as the sample size increases. No matter how the population distribution is shaped. i.e if we take some sample points from the distribution and calculate its mean then the distribution of those mean points will follow a normal/gaussian distribution no matter from which distribution we have taken the sample points.

There is one condition that the size of the sample must be greater than or equal to 30 for the CLT to hold. and the mean of the sample means approaches the population mean.

75. Explain the working principle of SVM.

A data set that is not separable in different classes in one plane may be separable in another plane. This is exactly the idea behind the SVM in this a low dimensional data is mapped to high dimensional data so, that it becomes separable in the different classes. A hyperplane is determined after mapping the data into a higher dimension which can separate the data into categories. SVM model can even learn non-linear boundaries with the objective that there should be as much margin as possible between the categories in which the data has been categorized. To perform this mapping different types of kernels are used like radial basis kernel, gaussian kernel, polynomial kernel, and many others.

76. What is the difference between the k-means and k-means++ algorithms?

The only difference between the two is in the way centroids are initialized. In the k-means algorithm, the centroids are initialized randomly from the given points. There is a drawback in this method that sometimes this random initialization leads to non-optimized clusters due to maybe initialization of two clusters close to each other.

To overcome this problem k-means++ algorithm was formed. In k-means++, The first centroid is selected randomly from the data points. The selection of subsequent centroids is based on their separation from the initial centroids. The probability of a point being selected as the next centroid is proportional to the squared distance between the point and the closest centroid that has already been selected. This guarantees that the centroids are evenly spread apart and lowers the possibility of convergence to less-than-ideal clusters. This helps the algorithm reach the global minima instead of getting stuck at some local minima. Read more about it here.

77. Explain some measures of similarity which are generally used in Machine learning.

Some of the most commonly used similarity measures are as follows:

• Cosine Similarity – By considering the two vectors in n – dimension we evaluate the cosine of the angle between the two. The range of this similarity measure varies from [-1, 1] where the value 1 represents that the two vectors are highly similar and -1 represents that the two vectors are completely different from each other.
• Euclidean or Manhattan Distance – These two values represent the distances between the two points in an n-dimensional plane. The only difference between the two is in the way the two are calculated.
• Jaccard Similarity – It is also known as IoU or Intersection over union it is widely used in the field of object detection to evaluate the overlap between the predicted bounding box and the ground truth bounding box.

78. What happens to the mean, median, and mode when your data distribution is right skewed and left skewed?

In the case of a left-skewed distribution also known as a positively skewed distribution mean is greater than the median which is greater than the mode. But in the case of left-skewed distribution, the scenario is completely reversed.

Right Skewed Distribution

Mode < Median < Mean

 

Left Skewed Distribution,

Mean <Median < Mode

79. Whether decision tree or random forest is more robust to the outliers.

Decision trees and random forests are both relatively robust to outliers. A random forest model is an ensemble of multiple decision trees so, the output of a random forest model is an aggregate of multiple decision trees.

So, when we average the results the chances of overfitting get reduced. Hence we can say that the random forest models are more robust to outliers.

80. What is the difference between L1 and L2 regularization? What is their significance?

L1 regularization: In L1 regularization also known as Lasso regularization in which we add the sum of absolute values of the weights of the model in the loss function. In L1 regularization weights for those features which are not at all important are penalized to zero so, in turn, we obtain feature selection by using the L1 regularization technique.

L2 regularization: In L2 regularization also known as Ridge regularization in which we add the square of the weights to the loss function. In both of these regularization methods, weights are penalized but there is a subtle difference between the objective they help to achieve.

In L2 regularization the weights are not penalized to 0 but they are near zero for irrelevant features. It is often used to prevent overfitting by shrinking the weights towards zero, especially when there are many features and the data is noisy.

81. What is a radial basis function? Explain its use.

RBF (radial basis function) is a real-valued function used in machine learning whose value only depends upon the input and fixed point called the center. The formula for the radial basis function is as follows:

Machine learning systems frequently use the RBF function for a variety of functions, including:

• RBF networks can be used to approximate complex functions. By training the network’s weights to suit a set of input-output pairs,
• RBF networks can be used for unsupervised learning to locate data groups. By treating the RBF centers as cluster centers,
• RBF networks can be used for classification tasks by training the network’s weights to divide inputs into groups based on how far from the RBF nodes they are.

It is one of the very famous kernels which is generally used in the SVM algorithm to map low dimensional data to a higher dimensional plane so, we can determine a boundary that can separate the classes in different regions of those planes with as much margin as possible.

82. Explain SMOTE method used to handle data imbalance.

The synthetic Minority Oversampling Technique is one of the methods which is used to handle the data imbalance problem in the dataset. In this method, we synthesized new data points using the existing ones from the minority classes by using linear interpolation. The advantage of using this method is that the model does not get trained on the same data. But the disadvantage of using this method is that it adds undesired noise to the dataset and can lead to a negative effect on the model’s performance.

83. Does the accuracy score always a good metric to measure the performance of a classification model?

No, there are times when we train our model on an imbalanced dataset the accuracy score is not a good metric to measure the performance of the model. In such cases, we use precision and recall to measure the performance of a classification model. Also, f1-score is another metric that can be used to measure performance but in the end, f1-score is also calculated using precision and recall as the f1-score is nothing but the harmonic mean of the precision and recall.

84. What is KNN Imputer?

We generally impute null values by the descriptive statistical measures of the data like mean, mode, or median but KNN Imputer is a more sophisticated method to fill the null values. A distance parameter is also used in this method which is also known as the k parameter. The work is somehow similar to the clustering algorithm. The missing value is imputed in reference to the neighborhood points of the missing values.

85. Explain the working procedure of the XGB model.

XGB model is an example of the ensemble technique of machine learning in this method weights are optimized in a sequential manner by passing them to the decision trees. After each pass, the weights become better and better as each tree tries to optimize the weights, and finally, we obtain the best weights for the problem at hand. Techniques like regularized gradient and mini-batch gradient descent have been used to implement this algorithm so, that it works in a very fast and optimized manner.

86. Explain some methods to handle missing values in that data.

Some of the methods to handle missing values are as follows:

• Removing the rows with null values may lead to the loss of some important information.
• Removing the column having null values if it has very less valuable information. it may lead to the loss of some important information.
• Imputing null values with descriptive statistical measures like mean, mode, and median.
• Using methods like KNN Imputer to impute the null values in a more sophisticated way.

87. What is the difference between k-means and the KNN algorithm?

k-means algorithm is one of the popular unsupervised machine learning algorithms which is used for clustering purposes. But the KNN is a model which is generally used for the classification task and is a supervised machine learning algorithm. The k-means algorithm helps us to label the data by forming clusters within the dataset.

88. What is Linear Discriminant Analysis?

LDA is a supervised machine learning dimensionality reduction technique because it uses target variables also for dimensionality reduction. It is commonly used for classification problems. The LDA mainly works on two objectives:

• Maximize the distance between the means of the two classes.
• Minimize the variation within each class.

89. How can we visualize high-dimensional data in 2-d?

One of the most common and effective methods is by using the t-SNE algorithm which is a short form for t-Distributed Stochastic Neighbor Embedding. This algorithm uses some non-linear complex methods to reduce the dimensionality of the given data. We can also use PCA or LDA to convert n-dimensional data to 2 – dimensional so, that we can plot it to get visuals for better analysis. But the difference between the PCA and t-SNE is that the former tries to preserve the variance of the dataset but the t-SNE tries to preserve the local similarities in the dataset.

90. What is the reason behind the curse of dimensionality?

As the dimensionality of the input data increases the amount of data required to generalize or learn the patterns present in the data increases. For the model, it becomes difficult to identify the pattern for every feature from the limited number of datasets or we can say that the weights are not optimized properly due to the high dimensionality of the data and the limited number of examples used to train the model. Due to this after a certain threshold for the dimensionality of the input data, we have to face the curse of dimensionality.

91. Whether the metric MAE or MSE or RMSE is more robust to the outliers.

Out of the above three metrics, MAE is robust to the outliers as compared to the MSE or RMSE. The main reason behind this is because of Squaring the error values. In the case of an outlier, the error value is already high and then we squared it which results in an explosion in the error values more than expected and creates misleading results for the gradient.

92. Why removing highly correlated features are considered a good practice?

When two features are highly correlated, they may provide similar information to the model, which may cause overfitting. If there are highly correlated features in the dataset then they unnecessarily increase the dimensionality of the feature space and sometimes create the problem of the curse of dimensionality. If the dimensionality of the feature space is high then the model training may take more time than expected, it will increase the complexity of the model and chances of error. This somehow also helps us to achieve data compression as the features have been removed without much loss of data.

93. What is the difference between the content-based and collaborative filtering algorithms of recommendation systems?

In a content-based recommendation system, similarities in the content and services are evaluated, and then by using these similarity measures from past data we recommend products to the user. But on the other hand in collaborative filtering, we recommend content and services based on the preferences of similar users. For example, if one user has taken A and B services in past and a new user has taken service A then service A will be recommended to him based on the other user’s preferences.

94. What is Packaging in machine learning?

Machine learning models are built and trained in a development environment, but they are deployed and used in a production environment, which often has different requirements and constraints. Model packaging ensures a machine learning model can be easily deployed and maintained in a production environment.

Proper model packaging ensures that a machine learning model is:

Easy to install: A well-packaged model should be straightforward to install, reducing the time and effort required for deployment.
Reproducible: Model packaging ensures that the model can be easily reproduced across different environments, providing consistent results.
Versioned: Keeping track of multiple model versions can be difficult, but model packaging makes it easier to version models, track changes, and roll back to previous versions if needed.
Documented: Good model packaging includes clear code documentation that helps others understand how to use and modify the model if required.

Natural Language Processing

1. What is NLP?

NLP stands for Natural Language Processing. The subfield of Artificial intelligence and computational linguistics deals with the interaction between computers and human languages. It involves developing algorithms, models, and techniques to enable machines to understand, interpret, and generate natural languages in the same way as a human does.

NLP encompasses a wide range of tasks, including language translation, sentiment analysis, text categorization, information extraction, speech recognition, and natural language understanding. NLP allows computers to extract meaning, develop insights, and communicate with humans in a more natural and intelligent manner by processing and analyzing textual input.

2. What are the main challenges in NLP?

The complexity and variety of human language create numerous difficult problems for the study of Natural Language Processing (NLP). The primary challenges in NLP are as follows:

• Semantics and Meaning: It is a difficult undertaking to accurately capture the meaning of words, phrases, and sentences. The semantics of the language, including word sense disambiguation, metaphorical language, idioms, and other linguistic phenomena, must be accurately represented and understood by NLP models.
• Ambiguity: Language is ambiguous by nature, with words and phrases sometimes having several meanings depending on context. Accurately resolving this ambiguity is a major difficulty for NLP systems.
• Contextual Understanding: Context is frequently used to interpret language. For NLP models to accurately interpret and produce meaningful replies, the context must be understood and used. Contextual difficulties include, for instance, comprehending referential statements and resolving pronouns to their antecedents.
• Language Diversity: NLP must deal with the world’s wide variety of languages and dialects, each with its own distinctive linguistic traits, lexicon, and grammar. The lack of resources and knowledge of low-resource languages complicates matters.
• Data Limitations and Bias: The availability of high-quality labelled data for training NLP models can be limited, especially for specific areas or languages. Furthermore, biases in training data might impair model performance and fairness, necessitating careful consideration and mitigation.
• Real-world Understanding: NLP models often fail to understand real-world knowledge and common sense, which humans are born with. Capturing and implementing this knowledge into NLP systems is a continuous problem.

3. What are the different tasks in NLP?

Natural Language Processing (NLP) includes a wide range of tasks involving understanding, processing, and creation of human language. Some of the most important tasks in NLP are as follows:

• Text Classification
• Named Entity Recognition (NER)
• Part-of-Speech Tagging (POS)
• Sentiment Analysis
• Language Modeling
• Machine Translation
• Chatbots
• Text Summarization
• Information Extraction
• Text Generation
• Speech Recognition

4. What do you mean by Corpus in NLP?

In NLP, a corpus is a huge collection of texts or documents. It is a structured dataset that acts as a sample of a specific language, domain, or issue. A corpus can include a variety of texts, including books, essays, web pages, and social media posts. Corpora are frequently developed and curated for specific research or NLP objectives. They serve as a foundation for developing language models, undertaking linguistic analysis, and gaining insights into language usage and patterns.

5. What do you mean by text augmentation in NLP and what are the different text augmentation techniques in NLP?

Text augmentation in NLP refers to the process that generates new or modified textual data from existing data in order to increase the diversity and quantity of training samples. Text augmentation techniques apply numerous alterations to the original text while keeping the underlying meaning.

Different text augmentation techniques in NLP include:

1. Synonym Replacement: Replacing words in the text with their synonyms to introduce variation while maintaining semantic similarity.
2. Random Insertion/Deletion: Randomly inserting or deleting words in the text to simulate noisy or incomplete data and enhance model robustness.
3. Word Swapping: Exchanging the positions of words within a sentence to generate alternative sentence structures.
4. Back translation: Translating the text into another language and then translating it back to the original language to introduce diverse phrasing and sentence constructions.
5. Random Masking: Masking or replacing random words in the text with a special token, akin to the approach used in masked language models like BERT.
6. Character-level Augmentation: Modifying individual characters in the text, such as adding noise, misspellings, or character substitutions, to simulate real-world variations.
7. Text Paraphrasing: Rewriting sentences or phrases using different words and sentence structures while preserving the original meaning.
8. Rule-based Generation: Applying linguistic rules to generate new data instances, such as using grammatical templates or syntactic transformations.

6. What are some common pre-processing techniques used in NLP?

Natural Language Processing (NLP) preprocessing refers to the set of processes and techniques used to prepare raw text input for analysis, modelling, or any other NLP tasks. The purpose of preprocessing is to clean and change text data so that it may be processed or analyzed later.

Preprocessing in NLP typically involves a series of steps, which may include:

• Tokenization
• Stop Word Removal
• Text Normalization
  • Lowercasing
  • Lemmatization
  • Stemming
  • Date and Time Normalization
• Removal of Special Characters and Punctuation
• Removing HTML Tags or Markup
• Spell Correction
• Sentence Segmentation

7. What is text normalization in NLP?

Text normalization, also known as text standardization, is the process of transforming text data into a standardized or normalized form It involves applying a variety of techniques to ensure consistency,  reduce variations, and simplify the representation of textual information.

The goal of text normalization is to make text more uniform and easier to process in Natural Language Processing (NLP) tasks. Some common techniques used in text normalization include:

• Lowercasing: Converting all text to lowercase to treat words with the same characters as identical and avoid duplication.
• Lemmatization: Converting words to their base or dictionary form, known as lemmas. For example, converting “running” to “run” or “better” to “good.”
• Stemming: Reducing words to their root form by removing suffixes or prefixes. For example, converting “playing” to “play” or “cats” to “cat.”
• Abbreviation Expansion: Expanding abbreviations or acronyms to their full forms. For example, converting “NLP” to “Natural Language Processing.”
• Numerical Normalization: Converting numerical digits to their written form or normalizing numerical representations. For example, converting “100” to “one hundred” or normalizing dates.
• Date and Time Normalization: Standardizing date and time formats to a consistent representation.

8. What is tokenization in NLP?

Tokenization is the process of breaking down text or string into smaller units called tokens. These tokens can be words, characters, or subwords depending on the specific applications. It is the fundamental step in many natural language processing tasks such as sentiment analysis, machine translation, and text generation. etc.

Some of the most common ways of tokenization are as follows:

• Sentence tokenization: In Sentence tokenizations, the text is broken down into individual sentences. This is one of the fundamental steps of tokenization.
• Word tokenization: In word tokenization, the text is simply broken down into words. This is one of the most common types of tokenization. It is typically done by splitting the text into spaces or punctuation marks.
• Subword tokenization: In subword tokenization, the text is broken down into subwords, which are the smaller part of words. Sometimes words are formed with more than one word, for example, Subword i.e Sub+ word, Here sub, and words have different meanings. When these two words are joined together, they form the new word “subword”, which means “a smaller unit of a word”. This is often done for tasks that require an understanding of the morphology of the text, such as stemming or lemmatization.
• Char-label tokenization: In Char-label tokenization, the text is broken down into individual characters. This is often used for tasks that require a more granular understanding of the text such as text generation, machine translations, etc.

9. What is NLTK and How it’s helpful in NLP?

NLTK stands for Natural Language Processing Toolkit. It is a suite of libraries and programs written in Python Language for symbolic and statistical natural language processing. It offers tokenization, stemming, lemmatization, POS tagging, Named Entity Recognization, parsing, semantic reasoning, and classification.

NLTK is a popular NLP library for Python. It is easy to use and has a wide range of features. It is also open-source, which means that it is free to use and modify.

10. What is stemming in NLP, and how is it different from lemmatization?

Stemming and lemmatization are two commonly used word normalization techniques in NLP, which aim to reduce the words to their base or root word. Both have similar goals but have different approaches.

In stemming, the word suffixes are removed using the heuristic or pattern-based rules regardless of the context of the parts of speech. The resulting stems may not always be actual dictionary words. Stemming algorithms are generally simpler and faster compared to lemmatization, making them suitable for certain applications with time or resource constraints.

In lemmatization, The root form of the word known as lemma, is determined by considering the word’s context and parts of speech. It uses linguistic knowledge and databases (e.g., wordnet) to transform words into their root form. In this case, the output lemma is a valid word as per the dictionary. For example, lemmatizing “running” and “runner” would result in “run.” Lemmatization provides better interpretability and can be more accurate for tasks that require meaningful word representations.

11. How does part-of-speech tagging work in NLP?

Part-of-speech tagging is the process of assigning a part-of-speech tag to each word in a sentence. The POS tags represent the syntactic information about the words and their roles within the sentence.

There are three main approaches for POS tagging:

• Rule-based POS tagging: It uses a set of handcrafted rules to determine the part of speech based on morphological, syntactic, and contextual patterns for each word in a sentence. For example, words ending with ‘-ing’ are likely to be a verb.
• Statistical POS tagging: The statistical model like Hidden Markov Model (HMMs) or Conditional Random Fields (CRFs) are trained on a large corpus of already tagged text. The model learns the probability of word sequences with their corresponding POS tags, and it can be further used for assigning each word to a most likely POS tag based on the context in which the word appears.
• Neural network POS tagging: The neural network-based model like RNN, LSTM, Bi-directional RNN, and transformer have given promising results in POS tagging by learning the patterns and representations of words and their context.

12. What is named entity recognition in NLP?

Named Entity Recognization (NER) is a task in natural language processing that is used to identify and classify the named entity in text. Named entity refers to real-world objects or concepts, such as persons, organizations, locations, dates, etc. NER is one of the challenging tasks in NLP because there are many different types of named entities, and they can be referred to in many different ways. The goal of NER is to extract and classify these named entities in order to offer structured data about the entities referenced in a given text.

The approach followed for Named Entity Recognization (NER) is the same as the POS tagging. The data used while training in NER is tagged with persons, organizations, locations, and dates.

13. What is parsing in NLP?

In NLP, parsing is defined as the process of determining the underlying structure of a sentence by breaking it down into constituent parts and determining the syntactic relationships between them according to formal grammar rules. The purpose of parsing is to understand the syntactic structure of a sentence, which allows for deeper learning of its meaning and encourages different downstream NLP tasks such as semantic analysis, information extraction, question answering, and machine translation. it is also known as syntax analysis or syntactic parsing.

The formal grammar rules used in parsing are typically based on Chomsky’s hierarchy. The simplest grammar in the Chomsky hierarchy is regular grammar, which can be used to describe the syntax of simple sentences. More complex grammar, such as context-free grammar and context-sensitive grammar, can be used to describe the syntax of more complex sentences.

14. What are the different types of parsing in NLP?

In natural language processing (NLP), there are several types of parsing algorithms used to analyze the grammatical structure of sentences. Here are some of the main types of parsing algorithms:

• Constituency Parsing: Constituency parsing in NLP tries to figure out a sentence’s hierarchical structure by breaking it into constituents based on a particular grammar. It generates valid constituent structures using context-free grammar. The parse tree that results represents the structure of the sentence, with the root node representing the complete sentence and internal nodes representing phrases. Constituency parsing techniques like as CKY, Earley, and chart parsing are often used for parsing. This approach is appropriate for tasks that need a thorough comprehension of sentence structure, such as semantic analysis and machine translation. When a complete understanding of sentence structure is required, constituency parsing, a classic parsing approach, is applied.
• Dependency Parsing: In NLP, dependency parsing identifies grammatical relationships between words in a sentence. It represents the sentence as a directed graph, with dependencies shown as labelled arcs. The graph emphasises subject-verb, noun-modifier, and object-preposition relationships. The head of a dependence governs the syntactic properties of another word. Dependency parsing, as opposed to constituency parsing, is helpful for languages with flexible word order. It allows for the explicit illustration of word-to-word relationships, resulting in a clear representation of grammatical structure.
• Top-down parsing: Top-down parsing starts at the root of the parse tree and iteratively breaks down the sentence into smaller and smaller parts until it reaches the leaves. This is a more natural technique for parsing sentences. However, because it requires a more complicated language, it may be more difficult to implement.
• Bottom-up parsing: Bottom-up parsing starts with the leaves of the parse tree and recursively builds up the tree from smaller and smaller constituents until it reaches the root. Although this method of parsing requires simpler grammar, it is frequently simpler to implement, even when it is less understandable.

15. What do you mean by vector space in NLP?

In natural language processing (NLP), A vector space is a mathematical vector where words or documents are represented by numerical vectors form. The word or document’s specific features or attributes are represented by one of the dimensions of the vector. Vector space models are used to convert text into numerical representations that machine learning algorithms can understand.

Vector spaces are generated using techniques such as word embeddings, bag-of-words, and term frequency-inverse document frequency (TF-IDF). These methods allow for the conversion of textual data into dense or sparse vectors in a high-dimensional space. Each dimension of the vector may indicate a different feature, such as the presence or absence of a word, word frequency, semantic meaning, or contextual information.

16. What is the bag-of-words model?

Bag of Words is a classical text representation technique in NLP that describes the occurrence of words within a document or not. It just keeps track of word counts and ignores the grammatical details and the word order.

Each document is transformed as a numerical vector, where each dimension corresponds to a unique word in the vocabulary. The value in each dimension of the vector represents the frequency, occurrence, or other measure of importance of that word in the document.

Let's consider two simple text documents:
Document 1: "I love apples."
Document 2: "I love mangoes too."

Step 1: Tokenization
Document 1 tokens: ["I", "love", "apples"]
Document 2 tokens: ["I", "love", "mangoes", "too"]

Step 2: Vocabulary Creation by collecting all unique words across the documents
Vocabulary: ["I", "love", "apples", "mangoes", "too"]
The vocabulary has five unique words, so each document vector will have five dimensions.

Step 3: Vectorization
Create numerical vectors for each document based on the vocabulary.
For Document 1:
- The dimension corresponding to "I" has a value of 1.
- The dimension corresponding to "love" has a value of 1.
- The dimension corresponding to "apples" has a value of 1.
- The dimensions corresponding to "mangoes" and "too" have values of 0 since they do not appear in Document 1.
Document 1 vector: [1, 1, 1, 0, 0]

For Document 2:
- The dimension corresponding to "I" has a value of 1.
- The dimension corresponding to "love" has a value of 1.
- The dimension corresponding to "mangoes" has a value of 1.
- The dimension corresponding to "apples" has a value of 0 since it does not appear in Document 2.
- The dimension corresponding to "too" has a value of 1.
Document 2 vector: [1, 1, 0, 1, 1]

The value in each dimension represents the occurrence or frequency of the corresponding word in the document. The BoW representation allows us to compare and analyze the documents based on their word frequencies.

17. Define the Bag of N-grams model in NLP.

The Bag of n-grams model is a modification of the standard bag-of-words (BoW) model in NLP. Instead of taking individual words to be the fundamental units of representation, the Bag of n-grams model considers contiguous sequences of n words, known as n-grams, to be the fundamental units of representation.

The Bag of n-grams model divides the text into n-grams, which can represent consecutive words or characters depending on the value of n. These n-grams are subsequently considered as features or tokens, similar to individual words in the BoW model.

The steps for creating a bag-of-n-grams model are as follows:

• The text is split or tokenized into individual words or characters.
• The tokenized text is used to construct N-grams of size n (sequences of n consecutive words or characters). If n is set to 1 known as uni-gram i.e. same as a bag of words, 2 i.e. bi-grams, and 3 i.e. tri-gram.
• A vocabulary is built by collecting all unique n-grams across the entire corpus.
• Similarly to the BoW approach, each document is represented as a numerical vector. The vector’s dimensions correspond to the vocabulary’s unique n-grams, and the value in each dimension denotes the frequency or occurrence of that n-gram in the document.

18. What is the term frequency-inverse document frequency (TF-IDF)?

Term frequency-inverse document frequency (TF-IDF) is a classical text representation technique in NLP that uses a statistical measure to evaluate the importance of a word in a document relative to a corpus of documents. It is a combination of two terms: term frequency (TF) and inverse document frequency (IDF).

• Term Frequency (TF): Term frequency measures how frequently a word appears in a document. it is the ratio of the number of occurrences of a term or word (t ) in a given document (d) to the total number of terms in a given document (d). A higher term frequency indicates that a word is more important within a specific document.
• Inverse Document Frequency (IDF): Inverse document frequency measures the rarity or uniqueness of a term across the entire corpus. It is calculated by taking the logarithm of the ratio of the total number of documents in the corpus to the number of documents containing the term. it down the weight of the terms, which frequently occur in the corpus, and up the weight of rare terms.

The TF-IDF score is calculated by multiplying the term frequency (TF) and inverse document frequency (IDF) values for each term in a document. The resulting score indicates the term’s importance in the document and corpus. Terms that appear frequently in a document but are uncommon in the corpus will have high TF-IDF scores, suggesting their importance in that specific document.

19. Explain the concept of cosine similarity and its importance in NLP.

The similarity between two vectors in a multi-dimensional space is measured using the cosine similarity metric. To determine how similar or unlike the vectors are to one another, it calculates the cosine of the angle between them.

In natural language processing (NLP), Cosine similarity is used to compare two vectors that represent text. The degree of similarity is calculated using the cosine of the angle between the document vectors. To compute the cosine similarity between two text document vectors, we often used the following procedures:

• Text Representation: Convert text documents into numerical vectors using approaches like bag-of-words, TF-IDF (Term Frequency-Inverse Document Frequency), or word embeddings like Word2Vec or GloVe.
• Vector Normalization: Normalize the document vectors to unit length. This normalization step ensures that the length or magnitude of the vectors does not affect the cosine similarity calculation.
• Cosine Similarity Calculation: Take the dot product of the normalised vectors and divide it by the product of the magnitudes of the vectors to obtain the cosine similarity.

Mathematically, the cosine similarity between two document vectors, �⃗   a and �⃗   b , can be expressed as:

Cosine Similarity(�⃗,�⃗)=�⃗⋅�⃗∣�⃗∣∣�⃗∣Cosine Similarity(a,b)=∣a∣∣b∣a⋅b​

Here,

• �⃗⋅�⃗   a⋅b is the dot product of vectors a and b
• |a| and |b| represent the Euclidean norms (magnitudes) of vectors a and b, respectively.

The resulting cosine similarity score ranges from -1 to 1, where 1 represents the highest similarity, 0 represents no similarity, and -1 represents the maximum dissimilarity between the documents.

20. What are the differences between rule-based, statistical-based and neural-based approaches in NLP?

Natural language processing (NLP) uses three distinct approaches to tackle language understanding and processing tasks: rule-based, statistical-based, and neural-based.

1. Rule-based Approach: Rule-based systems rely on predefined sets of linguistic rules and patterns to analyze and process language.
   • Linguistic Rules are manually crafted rules by human experts to define patterns or grammar structures.
   • The knowledge in rule-based systems is explicitly encoded in the rules, which may cover syntactic, semantic, or domain-specific information.
   • Rule-based systems offer high interpretability as the rules are explicitly defined and understandable by human experts.
   • These systems often require manual intervention and rule modifications to handle new language variations or domains.
2. Statistical-based Approach: Statistical-based systems utilize statistical algorithms and models to learn patterns and structures from large datasets.
   • By examining the data’s statistical patterns and relationships, these systems learn from training data.
   • Statistical models are more versatile than rule-based systems because they can train on relevant data from various topics and languages.
3. Neural-based Approach: Neural-based systems employ deep learning models, such as neural networks, to learn representations and patterns directly from raw text data.
   • Neural networks learn hierarchical representations of the input text, which enable them to capture complex language features and semantics.
   • Without explicit rule-making or feature engineering, these systems learn directly from data.
   • By training on huge and diverse datasets, neural networks are very versatile and can perform a wide range of NLP tasks.
   • In many NLP tasks, neural-based models have attained state-of-the-art performance, outperforming classic rule-based or statistical-based techniques.

21. What do you mean by Sequence in the Context of NLP?

A Sequence primarily refers to the sequence of elements that are analyzed or processed together. In NLP, a sequence may be a sequence of characters, a sequence of words or a sequence of sentences.

In general, sentences are often treated as sequences of words or tokens. Each word in the sentence is considered an element in the sequence. This sequential representation allows for the analysis and processing of sentences in a structured manner, where the order of words matters.

By considering sentences as sequences, NLP models can capture the contextual information and dependencies between words, enabling tasks such as part-of-speech tagging, named entity recognition, sentiment analysis, machine translation, and more.

22. What are the various types of machine learning algorithms used in NLP?

There are various types of machine learning algorithms that are often employed in natural language processing (NLP) tasks. Some of them are as follows:

• Naive Bayes: Naive Bayes is a probabilistic technique that is extensively used in NLP for text classification tasks. It computes the likelihood of a document belonging to a specific class based on the presence of words or features in the document.
• Support Vector Machines (SVM): SVM is a supervised learning method that can be used for text classification, sentiment analysis, and named entity recognition. Based on the given set of features, SVM finds a hyperplane that splits data points into various classes.
• Decision Trees: Decision trees are commonly used for tasks such as sentiment analysis, and information extraction. These algorithms build a tree-like model based on an order of decisions and feature conditions, which helps in making predictions or classifications.
• Random Forests: Random forests are a type of ensemble learning that combines multiple decision trees to improve accuracy and reduce overfitting.  They can be applied to the tasks like text classification, named entity recognition, and sentiment analysis.
• Recurrent Neural Networks (RNN): RNNs are a type of neural network architecture that are often used in sequence-based NLP tasks like language modelling, machine translation, and sentiment analysis. RNNs can capture temporal dependencies and context within a word sequence.
• Long Short-Term Memory (LSTM): LSTMs are a type of recurrent neural network that was developed to deal with the vanishing gradient problem of RNN. LSTMs are useful for capturing long-term dependencies in sequences, and they have been used in applications such as machine translation, named entity identification, and sentiment analysis.
• Transformer: Transformers are a relatively recent architecture that has gained significant attention in NLP. By exploiting self-attention processes to capture contextual relationships in text, transformers such as the BERT (Bidirectional Encoder Representations from Transformers) model have achieved state-of-the-art performance in a wide range of NLP tasks.

23. What is Sequence Labelling in NLP?

Sequence labelling is one of the fundamental NLP tasks in which, categorical labels are assigned to each individual element in a sequence. The sequence can represent various linguistic units such as words, characters, sentences, or paragraphs.

Sequence labelling in NLP includes the following tasks.

• Part-of-Speech Tagging (POS Tagging): In which part-of-speech tags (e.g., noun, verb, adjective) are assigned to each word in a sentence.
• Named Entity Recognition (NER): In which named entities like person names, locations, organizations, or dates are recognized and tagged in the sentences.
• Chunking: Words are organized into syntactic units or “chunks” based on their grammatical roles (for example, noun phrase, verb phrase).
• Semantic Role Labeling (SRL): In which, words or phrases in a sentence are labelled based on their semantic roles like Teacher, Doctor, Engineer, Lawyer etc
• Speech Tagging: In speech processing tasks such as speech recognition or phoneme classification, labels are assigned to phonetic units or acoustic segments.

Machine learning models like Conditional Random Fields (CRFs), Hidden Markov Models (HMMs), recurrent neural networks (RNNs), or transformers are used for sequence labelling tasks. These models learn from the labelled training data to make predictions on unseen data.

24.What is topic modelling in NLP?

Topic modelling is Natural Language Processing task used to discover hidden topics from large text documents. It is an unsupervised technique, which takes unlabeled text data as inputs and applies the probabilistic models that represent the probability of each document being a mixture of topics. For example, A document could have a 60% chance of being about neural networks, a 20% chance of being about Natural Language processing, and a 20% chance of being about anything else.

Where each topic will be distributed over words means each topic is a list of words, and each word has a probability associated with it. and the words that have the highest probabilities in a topic are the words that are most likely to be used to describe that topic. For example, the words like “neural”, “RNN”, and “architecture” are the keywords for neural networks and the words like ‘language”, and “sentiment” are the keywords for Natural Language processing.

There are a number of topic modelling algorithms but two of the most popular topic modelling algorithms are as follows:

• Latent Dirichlet Allocation (LDA): LDA is based on the idea that each text in the corpus is a mash-up of various topics and that each word in the document is derived from one of those topics. It is assumed that there is an unobservable (latent) set of topics and each document is generated by Topic Selection or Word Generation.
• Non-Negative Matrix Factorization (NMF): NMF is a matrix factorization technique that approximates the term-document matrix (where rows represent documents and columns represent words) into two non-negative matrices: one representing the topic-word relationships and the other the document-topic relationships. NMF aims to identify representative topics and weights for each document.

Topic modelling is especially effective for huge text collections when manually inspecting and categorising each document would be impracticable and time-consuming. We can acquire insights into the primary topics and structures of text data by using topic modelling, making it easier to organise, search, and analyse enormous amounts of unstructured text.

25. What is the GPT?

GPT stands for “Generative Pre-trained Transformer”. It refers to a collection of large language models created by OpenAI. It is trained on a massive dataset of text and code, which allows it to generate text, generate code, translate languages, and write many types of creative content, as well as answer questions in an informative manner. The GPT series includes various models, the most well-known and commonly utilised of which are the GPT-2 and GPT-3.

GPT models are built on the Transformer architecture, which allows them to efficiently capture long-term dependencies and contextual information in text. These models are pre-trained on a large corpus of text data from the internet, which enables them to learn the underlying patterns and structures of language.

26. What are word embeddings in NLP?

Word embeddings in NLP are defined as the dense, low-dimensional vector representations of words that capture semantic and contextual information about words in a language. It is trained using big text corpora through unsupervised or supervised methods to represent words in a numerical format that can be processed by machine learning models.

The main goal of Word embeddings is to capture relationships and similarities between words by representing them as dense vectors in a continuous vector space. These vector representations are acquired using the distributional hypothesis, which states that words with similar meanings tend to occur in similar contexts. Some of the popular pre-trained word embeddings are Word2Vec, GloVe (Global Vectors for Word Representation), or FastText. The advantages of word embedding over the traditional text vectorization technique are as follows:

• It can capture the Semantic Similarity between the words
• It is capable of capturing syntactic links between words. Vector operations such as “king” – “man” + “woman” may produce a vector similar to the vector for “queen,” capturing the gender analogy.
• Compared to one-shot encoding, it has reduced the dimensionality of word representations. Instead of high-dimensional sparse vectors, word embeddings typically have a fixed length and represent words as dense vectors.
• It can be generalized to represent words that they have not been trained on i.e. out-of-vocabulary words. This is done by using the learned word associations to place new words in the vector space near words that they are semantically or syntactically similar to.

27. What are the various algorithms used for training word embeddings?

There are various approaches that are typically used for training word embeddings, which are dense vector representations of words in a continuous vector space. Some of the popular word embedding algorithms are as follows:

• Word2Vec: Word2vec is a common approach for generating vector representations of words that reflect their meaning and relationships. Word2vec learns embeddings using a shallow neural network and follows two approaches: CBOW and Skip-gram
  • CBOW (Continuous Bag-of-Words) predicts a target word based on its context words.
  • Skip-gram predicts context words given a target word.
• GloVe: GloVe (Global Vectors for Word Representation) is a word embedding model that is similar to Word2vec. GloVe, on the other hand, uses  objective function that constructs a co-occurrence matrix based on the statistics of word co-occurrences in a large corpus. The co-occurrence matrix is a square matrix where each entry represents the number of times two words co-occur in a window of a certain size. GloVe then performs matrix factorization on the co-occurrence matrix. Matrix factorization is a technique for finding a low-dimensional representation of a high-dimensional matrix. In the case of GloVe, the low-dimensional representation is a vector representation for each word in the corpus. The word embeddings are learned by minimizing a loss function that measures the difference between the predicted co-occurrence probabilities and the actual co-occurrence probabilities. This makes GloVe more robust to noise and less sensitive to the order of words in a sentence.
• FastText: FastText is a Word2vec extension that includes subword information. It represents words as bags of character n-grams, allowing it to handle out-of-vocabulary terms and capture morphological information. During training, FastText considers subword information as well as word context..
• ELMo: ELMo is a deeply contextualised word embedding model that generates context-dependent word representations. It generates word embeddings that capture both semantic and syntactic information based on the context of the word using bidirectional language models.
• BERT: A transformer-based model called BERT (Bidirectional Encoder Representations from Transformers) learns contextualised word embeddings. BERT is trained on a large corpus by anticipating masked terms inside a sentence and gaining knowledge about the bidirectional context. The generated embeddings achieve state-of-the-art performance in many NLP tasks and capture extensive contextual information.

28. How to handle out-of-vocabulary (OOV) words in NLP?

OOV words are words that are missing in a language model’s vocabulary or the training data it was trained on. Here are a few approaches to handling OOV words in NLP:

1. Character-level models: Character-level models can be used in place of word-level representations. In this method, words are broken down into individual characters, and the model learns representations based on character sequences. As a result, the model can handle OOV words since it can generalize from known character patterns.
2. Subword tokenization: Byte-Pair Encoding (BPE) and WordPiece are two subword tokenization algorithms that divide words into smaller subword units based on their frequency in the training data. This method enables the model to handle OOV words by representing them as a combination of subwords that it comes across during training.
3. Unknown token: Use a special token, frequently referred to as an “unknown” token or “UNK,” to represent any OOV term that appears during inference. Every time the model comes across an OOV term, it replaces it with the unidentified token and keeps processing. The model is still able to generate relevant output even though this technique doesn’t explicitly define the meaning of the OOV word.
4. External knowledge: When dealing with OOV terms, using external knowledge resources, like a knowledge graph or an external dictionary, can be helpful. We need to try to look up a word’s definition or relevant information in the external knowledge source when we come across an OOV word.
5. Fine-tuning: We can fine-tune using the pre-trained language model with domain-specific or task-specific data that includes OOV words. By incorporating OOV words in the fine-tuning process, we expose the model to these words and increase its capacity to handle them.

29. What is the difference between a word-level and character-level language model?

The main difference between a word-level and a character-level language model is how text is represented. A character-level language model represents text as a sequence of characters, whereas a word-level language model represents text as a sequence of words.

Word-level language models are often easier to interpret and more efficient to train. They are, however, less accurate than character-level language models because they cannot capture the intricacies of the text that are stored in the character order. Character-level language models are more accurate than word-level language models, but they are more complex to train and interpret. They are also more sensitive to noise in the text, as a slight alteration in a character can have a large impact on the meaning of the text.

The key differences between word-level and character-level language models are:

	Word-level	Character-level
Text representation	Sequence of words	Sequence of characters
Interpretability	Easier to interpret	More difficult to interpret
Sensitivity to noise	Less sensitive	More sensitive
Vocabulary	Fixed vocabulary of words	No predefined vocabulary
Out-of-vocabulary (OOV) handling	Struggles with OOV words	Naturally handles OOV words
Generalization	Captures semantic relationships between words	Better at handling morphological details
Training complexity	Smaller input/output space, less computationally intensive	Larger input/output space, more computationally intensive
Applications	Well-suited for tasks requiring word-level understanding	Suitable for tasks requiring fine-grained details or morphological variations

30. What is word sense disambiguation?

The task of determining which sense of a word is intended in a given context is known as word sense disambiguation (WSD). This is a challenging task because many words have several meanings that can only be determined by considering the context in which the word is used.

For example, the word “bank” can be used to refer to a variety of things, including “a financial institution,” “a riverbank,” and “a slope.” The term “bank” in the sentence “I went to the bank to deposit my money” should be understood to mean “a financial institution.” This is so because the sentence’s context implies that the speaker is on their way to a location where they can deposit money.

31. What is co-reference resolution?

Co-reference resolution is a natural language processing (NLP) task that involves identifying all expressions in a text that refer to the same entity. In other words, it tries to determine whether words or phrases in a text, typically pronouns or noun phrases, correspond to the same real-world thing. For example, the pronoun “he” in the sentence “Pawan Gunjan has compiled this article, He had done lots of research on Various NLP interview questions” refers to Pawan Gunjan himself. Co-reference resolution automatically identifies such linkages and establishes that “He” refers to “Pawan Gunjan” in all instances.

Co-reference resolution is used in information extraction, question answering, summarization, and dialogue systems because it helps to generate more accurate and context-aware representations of text data. It is an important part of systems that require a more in-depth understanding of the relationships between entities in large text corpora.

32.What is information extraction?

Information extraction is a natural language processing task used to extract specific pieces of information like names, dates, locations, and relationships etc from unstructured or semi-structured texts.

Natural language is often ambiguous and can be interpreted in a variety of ways, which makes IE a difficult process. Some of the common techniques used for information extraction include:

• Named entity recognition (NER): In NER, named entities like people, organizations, locations, dates, or other specific categories are recognized from the text documents. For NER problems, a variety of machine learning techniques, including conditional random fields (CRF), support vector machines (SVM), and deep learning models, are frequently used.
• Relationship extraction: In relationship extraction, the connections between the stated text are identified. I figure out the relations different kinds of relationships between various things like “is working at”, “lives in” etc.
• Coreference resolution: Coreference resolution is the task of identifying the referents of pronouns and other anaphoric expressions in the text. A coreference resolution system, for example, might be able to figure out that the pronoun “he” in a sentence relates to the person “John” who was named earlier in the text.
• Deep Learning-based Approaches: To perform information extraction tasks, deep learning models such as recurrent neural networks (RNNs), transformer-based architectures (e.g., BERT, GPT), and deep neural networks have been used. These models can learn patterns and representations from data automatically, allowing them to manage complicated and diverse textual material.

33. What is the Hidden Markov Model, and How it’s helpful in NLP tasks?

Hidden Markov Model is a probabilistic model based on the Markov Chain Rule used for modelling sequential data like characters, words, and sentences by computing the probability distribution of sequences.

Markov chain uses the Markov assumptions which state that the probabilities future state of the system only depends on its present state, not on any past state of the system. This assumption simplifies the modelling process by reducing the amount of information needed to predict future states.

The underlying process in an HMM is represented by a set of hidden states that are not directly observable. Based on the hidden states, the observed data, such as characters, words, or phrases, are generated.

Hidden Markov Models consist of two key components:

1. Transition Probabilities: The transition probabilities in Hidden Markov Models(HMMs) represents the likelihood of moving from one hidden state to another. It captures the dependencies or relationships between adjacent states in the sequence. In part-of-speech tagging, for example, the HMM’s hidden states represent distinct part-of-speech tags, and the transition probabilities indicate the likelihood of transitioning from one part-of-speech tag to another.
2. Emission Probabilities: In HMMs, emission probabilities define the likelihood of observing specific symbols (characters, words, etc.) given a particular hidden state. The link between the hidden states and the observable symbols is encoded by these probabilities.
3. Emission probabilities are often used in NLP to represent the relationship between words and linguistic features such as part-of-speech tags or other linguistic variables. The HMM captures the likelihood of generating an observable symbol (e.g., word) from a specific hidden state (e.g., part-of-speech tag) by calculating the emission probabilities.

Hidden Markov Models (HMMs) estimate transition and emission probabilities from labelled data using approaches such as the Baum-Welch algorithm. Inference algorithms like Viterbi and Forward-Backward are used to determine the most likely sequence of hidden states given observed symbols. HMMs are used to represent sequential data and have been implemented in NLP applications such as part-of-speech tagging. However, advanced models, such as CRFs and neural networks, frequently beat HMMs due to their flexibility and ability to capture richer dependencies.

34. What is the conditional random field (CRF) model in NLP?

Conditional Random Fields are a probabilistic graphical model that is designed to predict the sequence of labels for a given sequence of observations. It is well-suited for prediction tasks in which contextual information or dependencies among neighbouring elements are crucial.

CRFs are an extension of Hidden Markov Models (HMMs) that allow for the modelling of more complex relationships between labels in a sequence. It is specifically designed to capture dependencies between non-consecutive labels, whereas HMMs presume a Markov property in which the current state is only dependent on the past state. This makes CRFs more adaptable and suitable for capturing long-term dependencies and complicated label interactions.

In a CRF model, the labels and observations are represented as a graph. The nodes in the graph represent the labels, and the edges represent the dependencies between the labels. The model assigns weights to features that capture relevant information about the observations and labels.

During training, the CRF model learns the weights by maximizing the conditional log-likelihood of the labelled training data. This process involves optimization algorithms such as gradient descent or the iterative scaling algorithm.

During inference, given an input sequence, the CRF model calculates the conditional probabilities of different label sequences. Algorithms like the Viterbi algorithm efficiently find the most likely label sequence based on these probabilities.

CRFs have demonstrated high performance in a variety of sequence labelling tasks like named entity identification, part-of-speech tagging, and others.

35. What is a recurrent neural network (RNN)?

Recurrent Neural Networks are the type of artificial neural network that is specifically built to work with sequential or time series data. It is utilised in natural language processing activities such as language translation, speech recognition, sentiment analysis, natural language production, summary writing, and so on. It differs from feedforward neural networks in that the input data in RNN does not only flow in a single direction but also has a loop or cycle inside its design that has “memory” that preserves information over time. As a result, the RNN can handle data where context is critical, such as natural languages.

RNNs work by analysing input sequences one element at a time while keeping track in a hidden state that provides a summary of the sequence’s previous elements. At each time step, the hidden state is updated based on the current input and the prior hidden state. RNNs can thus capture the temporal connections between sequence items and use that knowledge to produce predictions.

36. How does the Backpropagation through time work in RNN?

Backpropagation through time(BPTT) propagates gradient information across the RNN’s recurrent connections over a sequence of input data. Let’s understand step by step process for BPTT.

1. Forward Pass: The input sequence is fed into the RNN one element at a time, starting from the first element. Each input element is processed through the recurrent connections, and the hidden state of the RNN is updated.
2. Hidden State Sequence: The hidden state of the RNN is maintained and carried over from one time step to the next. It contains information about the previous inputs and hidden states in the sequence.
3. Output Calculation: The updated hidden state is used to compute the output at each time step.
4. Loss Calculation: At the end of the sequence, the predicted output is compared to the target output, and a loss value is calculated using a suitable loss function, such as mean squared error or cross-entropy loss.
5. Backpropagation: The loss is then backpropagated through time, starting from the last time step and moving backwards in time. The gradients of the loss with respect to the parameters of the RNN are calculated at each time step.
6. Weight Update: The gradients are accumulated over the entire sequence, and the weights of the RNN are updated using an optimization algorithm such as gradient descent or its variants.
7. Repeat: The process is repeated for a specified number of epochs or until convergence, during this the training data is iterated through several times.

During the backpropagation step, the gradients at each time step are obtained and used to update the weights of the recurrent connections. This accumulation of gradients over numerous time steps allows the RNN to learn and capture dependencies and patterns in sequential data.

37. What are the limitations of a standard RNN?

Standard RNNs (Recurrent Neural Networks) have several limitations that can make them unsuitable for certain applications:

1. Vanishing Gradient Problem: Standard RNNs are vulnerable to the vanishing gradient problem, in which gradients decrease exponentially as they propagate backwards through time. Because of this issue, it is difficult for the network to capture and transmit long-term dependencies across multiple time steps during training.
2. Exploding Gradient Problem: RNNs, on the other hand, can suffer from the expanding gradient problem, in which gradients get exceedingly big and cause unstable training. This issue can cause the network to converge slowly or fail to converge at all.
3. Short-Term Memory: Standard RNNs have limited memory and fail to remember information from previous time steps. Because of this limitation, they have difficulty capturing long-term dependencies in sequences, limiting their ability to model complicated relationships that span a significant number of time steps.

38. What is a long short-term memory (LSTM) network?

A Long Short-Term Memory (LSTM) network is a type of recurrent neural network (RNN) architecture that is designed to solve the vanishing gradient problem and capture long-term dependencies in sequential data. LSTM networks are particularly effective in tasks that involve processing and understanding sequential data, such as natural language processing and speech recognition.

The key idea behind LSTMs is the integration of a memory cell, which acts as a memory unit capable of retaining information for an extended period. The memory cell is controlled by three gates: the input gate, the forget gate, and the output gate.

The input gate controls how much new information should be stored in the memory cell. The forget gate determines which information from the memory cell should be destroyed or forgotten. The output gate controls how much information is output from the memory cell to the next time step. These gates are controlled by activation functions, which are commonly sigmoid and tanh functions, and allow the LSTM to selectively update, forget, and output data from the memory cell.

39. What is the GRU model in NLP?

The Gated Recurrent Unit (GRU) model is a type of recurrent neural network (RNN) architecture that has been widely used in natural language processing (NLP) tasks. It is designed to address the vanishing gradient problem and capture long-term dependencies in sequential data.

GRU is similar to LSTM in that it incorporates gating mechanisms, but it has a simplified architecture with fewer gates, making it computationally more efficient and easier to train. The GRU model consists of the following components:

1. Hidden State: The hidden state ℎ�−1  ht−1​ in GRU represents the learned representation or memory of the input sequence up to the current time step. It retains and passes information from the past to the present.
2. Update Gate: The update gate in GRU controls the flow of information from the past hidden state to the current time step. It determines how much of the previous information should be retained and how much new information should be incorporated.
3. Reset Gate: The reset gate in GRU determines how much of the past information should be discarded or forgotten. It helps in removing irrelevant information from the previous hidden state.
4. Candidate Activation: The candidate activation represents the new information to be added to the hidden state ℎ�‘  ht‘​ . It is computed based on the current input and a transformed version of the previous hidden state using the reset gate.

GRU models have been effective in NLP applications like language modelling, sentiment analysis, machine translation, and text generation. They are particularly useful in situations when it is essential to capture long-term dependencies and understand the context. Due to its simplicity and computational efficiency, GRU makes it a popular choice in NLP research and applications.

40. What is the sequence-to-sequence (Seq2Seq) model in NLP?

Sequence-to-sequence (Seq2Seq) is a type of neural network that is used for natural language processing (NLP) tasks. It is a type of recurrent neural network (RNN) that can learn long-term word relationships. This makes it ideal for tasks like machine translation, text summarization, and question answering.

The model is composed of two major parts: an encoder and a decoder. Here’s how the Seq2Seq model works:

1. Encoder: The encoder transforms the input sequence, such as a sentence in the source language, into a fixed-length vector representation known as the “context vector” or “thought vector”. To capture sequential information from the input, the encoder commonly employs recurrent neural networks (RNNs) such as Long Short-Term Memory (LSTM) or Gated Recurrent Units (GRU).
2. Context Vector: The encoder’s context vector acts as a summary or representation of the input sequence. It encodes the meaning and important information from the input sequence into a fixed-size vector, regardless of the length of the input.
3. Decoder: The decoder uses the encoder’s context vector to build the output sequence, which could be a translation or a summarised version. It is another RNN-based network that creates the output sequence one token at a time. At each step, the decoder can be conditioned on the context vector, which serves as an initial hidden state.

During training, the decoder is fed ground truth tokens from the target sequence at each step. Backpropagation through time (BPTT) is a technique commonly used to train Seq2Seq models. The model is optimized to minimize the difference between the predicted output sequence and the actual target sequence.

The Seq2Seq model is used during prediction or generation to construct the output sequence word by word, with each predicted word given back into the model as input for the subsequent step. The process is repeated until either an end-of-sequence token or a predetermined maximum length is achieved.

41. How does the attention mechanism helpful in NLP?

An attention mechanism is a kind of neural network that uses an additional attention layer within an Encoder-Decoder neural network that enables the model to focus on specific parts of the input while performing a task. It achieves this by dynamically assigning weights to different elements in the input, indicating their relative importance or relevance. This selective attention allows the model to focus on relevant information, capture dependencies, and analyze relationships within the data.

The attention mechanism is particularly valuable in tasks involving sequential or structured data, such as natural language processing or computer vision, where long-term dependencies and contextual information are crucial for achieving high performance. By allowing the model to selectively attend to important features or contexts, it improves the model’s ability to handle complex relationships and dependencies in the data, leading to better overall performance in various tasks.

42. What is the Transformer model?

Transformer is one of the fundamental models in NLP based on the attention mechanism, which allows it to capture long-range dependencies in sequences more effectively than traditional recurrent neural networks (RNNs). It has given state-of-the-art results in various NLP tasks like word embedding, machine translation, text summarization, question answering etc.

Some of the key advantages of using a Transformer are as follows:

• Parallelization: The self-attention mechanism allows the model to process words in parallel, which makes it significantly faster to train compared to sequential models like RNNs.
• Long-Range Dependencies: The attention mechanism enables the Transformer to effectively capture long-range dependencies in sequences, which makes it suitable for tasks where long-term context is essential.
• State-of-the-Art Performance: Transformer-based models have achieved state-of-the-art performance in various NLP tasks, such as machine translation, language modelling, text generation, and sentiment analysis.

The key components of the Transformer model are as follows:

• Self-Attention Mechanism:
• Encoder-Decoder Network:
• Multi-head Attention:
• Positional Encoding
• Feed-Forward Neural Networks
• Layer Normalization and Residual Connections

43. What is the role of the self-attention mechanism in Transformers?

The self-attention mechanism is a powerful tool that allows the Transformer model to capture long-range dependencies in sequences. It allows each word in the input sequence to attend to all other words in the same sequence, and the model learns to assign weights to each word based on its relevance to the others. This enables the model to capture both short-term and long-term dependencies, which is critical for many NLP applications.

44. What is the purpose of the multi-head attention mechanism in Transformers?

The purpose of the multi-head attention mechanism in Transformers is to allow the model to recognize different types of correlations and patterns in the input sequence. In both the encoder and decoder, the Transformer model uses multiple attention heads. This enables the model to recognise different types of correlations and patterns in the input sequence. Each attention head learns to pay attention to different parts of the input, allowing the model to capture a wide range of characteristics and dependencies.

The multi-head attention mechanism helps the model in learning richer and more contextually relevant representations, resulting in improved performance on a variety of natural language processing (NLP) tasks.

45. What are positional encodings in Transformers, and why are they necessary?

The transformer model processes the input sequence in parallel, so that lacks the inherent understanding of word order like the sequential model recurrent neural networks (RNNs), LSTM possess. So, that. it requires a method to express the positional information explicitly.

Positional encoding is applied to the input embeddings to offer this positional information like the relative or absolute position of each word in the sequence to the model. These encodings are typically learnt and can take several forms, including sine and cosine functions or learned embeddings. This enables the model to learn the order of the words in the sequence, which is critical for many NLP tasks.

46. Describe the architecture of the Transformer model.

The architecture of the Transformer model is based on self-attention and feed-forward neural network concepts. It is made up of an encoder and a decoder, both of which are composed of multiple layers, each containing self-attention and feed-forward sub-layers. The model’s design encourages parallelization, resulting in more efficient training and improved performance on tasks involving sequential data, such as natural language processing (NLP) tasks.

The architecture can be described in depth below:

1. Encoder:
   • Input Embeddings: The encoder takes an input sequence of tokens (e.g., words) as input and transforms each token into a vector representation known as an embedding. Positional encoding is used in these embeddings to preserve the order of the words in the sequence.
   • Self-Attention Layers: An encoder consists of multiple self-attention layers and each self-attention layer is used to capture relationships and dependencies between words in the sequence.
   • Feed-Forward Layers: After the self-attention step, the output representations of the self-attention layer are fed into a feed-forward neural network. This network applies the non-linear transformations to each word’s contextualised representation independently.
   • Layer Normalization and Residual Connections: Residual connections and layer normalisation are used to back up the self-attention and feed-forward layers. The residual connections in deep networks help to mitigate the vanishing gradient problem, and layer normalisation stabilises the training process.
2. Decoder:
   • Input Embeddings: Similar to the encoder, the decoder takes an input sequence and transforms each token into embeddings with positional encoding.
   • Masked Self-Attention: Unlike the encoder, the decoder uses masked self-attention in the self-attention layers. This masking ensures that the decoder can only attend to places before the current word during training, preventing the model from seeing future tokens during generation.
   • Cross-Attention Layers: Cross-attention layers in the decoder allow it to attend to the encoder’s output, which enables the model to use information from the input sequence during output sequence generation.
   • Feed-Forward Layers: Similar to the encoder, the decoder’s self-attention output passes through feed-forward neural networks.
   • Layer Normalization and Residual Connections: The decoder also includes residual connections and layer normalization to help in training and improve model stability.
3. Final Output Layer:
   • Softmax Layer: The final output layer is a softmax layer that transforms the decoder’s representations into probability distributions over the vocabulary. This enables the model to predict the most likely token for each position in the output sequence.

Overall, the Transformer’s architecture enables it to successfully handle long-range dependencies in sequences and execute parallel computations, making it highly efficient and powerful for a variety of sequence-to-sequence tasks. The model has been successfully used for machine translation, language modelling, text generation, question answering, and a variety of other NLP tasks, with state-of-the-art results.

47. What is the difference between a generative and discriminative model in NLP?

Both generative and discriminative models are the types of machine learning models used for different purposes in the field of natural language processing (NLP).

Generative models are trained to generate new data that is similar to the data that was used to train them.  For example, a generative model could be trained on a dataset of text and code and then used to generate new text or code that is similar to the text and code in the dataset. Generative models are often used for tasks such as text generation, machine translation, and creative writing.

Discriminative models are trained to recognise different types of data. A discriminative model. For example, a discriminative model could be trained on a dataset of labelled text and then used to classify new text as either spam or ham. Discriminative models are often used for tasks such as text classification, sentiment analysis, and question answering.

The key differences between generative and discriminative models in NLP are as follows:

	Generative Models	Discriminative Models
Purpose	Generate new data that is similar to the training data.	Distinguish between different classes or categories of data.
Training	Learn the joint probability distribution of input and output data to generate new samples.	Learn the conditional probability distribution of the output labels given the input data.
Examples	Text generation, machine translation, creative writing, Chatbots, text summarization, and language modelling.	Text classification, sentiment analysis, and named entity recognition.

48. What is machine translation, and how does it is performed?

Machine translation is the process of automatically translating text or speech from one language to another using a computer or machine learning model.

There are three techniques for machine translation:

• Rule-based machine translation (RBMT): RBMT systems use a set of rules to translate text from one language to another.
• Statistical machine translation (SMT): SMT systems use statistical models to calculate the probability of a given translation being correct.
• Neural machine translation (NMT): Neural machine translation (NMT) is a recent technique of machine translation have been proven to be more accurate than RBMT and SMT systems, In recent years, neural machine translation (NMT), powered by deep learning models such as the Transformer, are becoming increasingly popular.

49. What is the BLEU score?

BLEU stands for “Bilingual Evaluation Understudy”. It is a metric invented by IBM in 2001 for evaluating the quality of a machine translation. It measures the similarity between machine-generated translations with the professional human translation. It was one of the first metrics whose results are very much correlated with human judgement.

The BLEU score is measured by comparing the n-grams (sequences of n words) in the machine-translated text to the n-grams in the reference text. The higher BLEU Score signifies, that the machine-translated text is more similar to the reference text.

The BLEU (Bilingual Evaluation Understudy) score is calculated using n-gram precision and a brevity penalty.

• N-gram Precision: The n-gram precision is the ratio of matching n-grams in the machine-generated translation to the total number of n-grams in the reference translation. The number of unigrams, bigrams, trigrams, and four-grams (i=1,…,4) that coincide with their n-gram counterpart in the reference translations is measured by the n-gram overlap.
  precision�=Count of matching n-gramscount of all n-grams in the machine translation  precisioni​=count of all n-grams in the machine translationCount of matching n-grams​ 
  For BLEU score precision�  precisioni​ is calculated for the I ranging (1 to N). Usually, the N value will be up to 4.
• Brevity Penalty: Brevity Penalty measures the length difference between machine-generated translations and reference translations. While finding the BLEU score, It penalizes the machine-generated translations if that is found too short compared to the reference translation’s length with exponential decay.
  brevity-penalty=min⁡(1,exp⁡(1−Reference lengthMachine translation length)))brevity-penalty=min(1,exp(1−Machine translation length)Reference length​))
• BLEU Score: The BLEU score is calculated by taking the geometric mean of the individual n-gram precisions and then adjusting it with the brevity penalty.
  BLEU=brevity-penalty×exp⁡[∑�=1�log⁡(precision�)�]=brevity-penalty×exp⁡[log⁡(∏�=1�precision�)�]=brevity-penalty×(∏�=1�precision�)1�  BLEU​=brevity-penalty×exp[N∑i=1N​log(precisioni​)​]=brevity-penalty×exp​Nlog(∏i=1N​precisioni​)​​=brevity-penalty×(i=1∏N​precisioni​)N1​​ 
  Here, N is the maximum n-gram size, (usually 4).

The BLEU score goes from 0 to 1, with higher values indicating better translation quality and 1 signifying a perfect match to the reference translation

50. List out the popular NLP task and their corresponding evaluation metrics.

Natural Language Processing (NLP) involves a wide range of tasks, each with its own set of objectives and evaluation criteria. Below is a list of common NLP tasks along with some typical evaluation metrics used to assess their performance:

Natural Language Processing(NLP) Tasks	Evaluation Metric
Part-of-Speech Tagging (POS Tagging) or Named Entity Recognition (NER)	Accuracy, F1-score, Precision, Recall
Dependency Parsing	UAS (Unlabeled Attachment Score), LAS (Labeled Attachment Score)
Coreference resolution	B-CUBED, MUC, CEAF
Text Classification or Sentiment Analysis	Accuracy, F1-score, Precision, Recall
Machine Translation	BLEU (Bilingual Evaluation Understudy), METEOR (Metric for Evaluation of Translation with Explicit Ordering)
Text Summarization	ROUGE (Recall-Oriented Understudy for Gisting Evaluation), BLEU
Question Answering	F1-score, Precision, Recall, MRR(Mean Reciprocal Rank)
Text Generation	Human evaluation (subjective assessment), perplexity (for language models)
Information Retrieval	Precision, Recall, F1-score, Mean Average Precision (MAP)
Natural language inference (NLI)	Accuracy, precision, recall, F1-score, Matthews correlation coefficient (MCC)
Topic Modeling	Coherence Score, Perplexity
Speech Recognition	Word Error Rate (WER)
Speech Synthesis (Text-to-Speech)	Mean Opinion Score (MOS)

The brief explanations of each of the evaluation metrics are as follows:

• Accuracy: Accuracy is the percentage of predictions that are correct.
• Precision: Precision is the percentage of correct predictions out of all the predictions that were made.
• Recall: Recall is the percentage of correct predictions out of all the positive cases.
• F1-score: F1-score is the harmonic mean of precision and recall.
• MAP(Mean Average Precision): MAP computes the average precision for each query and then averages those precisions over all queries.
• MUC(Mention-based Understudy for Coreference): MUC is a metric for coreference resolution that measures the number of mentions that are correctly identified and linked.
• B-CUBED: B-cubed is a metric for coreference resolution that measures the number of mentions that are correctly identified, linked, and ordered.
• CEAF: CEAF is a metric for coreference resolution that measures the similarity between the predicted coreference chains and the gold standard coreference chains.
• ROC AUC: ROC AUC is a metric for binary classification that measures the area under the receiver operating characteristic curve.
• MRR: MRR is a metric for question answering that measures the mean reciprocal rank of the top-k-ranked documents.
• Perplexity: Perplexity is a language model evaluation metric. It assesses how well a linguistic model predicts a sample or test set of previously unseen data. Lower perplexity values suggest that the language model is more predictive.
• BLEU: BLEU is a metric for machine translation that measures the n-gram overlap between the predicted translation and the gold standard translation.
• METEOR: METEOR is a metric for machine translation that measures the overlap between the predicted translation and the gold standard translation, taking into account synonyms and stemming.
• WER(Word Error Rate): WER is a metric for machine translation that measures the word error rate of the predicted translation.
• MCC: MCC is a metric for natural language inference that measures the Matthews correlation coefficient between the predicted labels and the gold standard labels.
• ROUGE: ROUGE is a metric for text summarization that measures the overlap between the predicted summary and the gold standard summary, taking into account n-grams and synonyms.
• Human Evaluation (Subjective Assessment): Human experts or crowd-sourced workers are asked to submit their comments, evaluations, or rankings on many elements of the NLP task’s performance in this technique.

51. What do you understand by Natural Language Processing?

Natural Language Processing is a field of computer science that deals with communication between computer systems and humans. It is a technique used in Artificial Intelligence and Machine Learning. It is used to create automated software that helps understand human-spoken languages to extract useful information from the data. Techniques in NLP allow computer systems to process and interpret data in the form of natural languages.

52. List any two real-life applications of Natural Language Processing.

Two real-life applications of Natural Language Processing are as follows:

1. Google Translate: Google Translate is one of the famous applications of Natural Language Processing. It helps convert written or spoken sentences into any language. Also, we can find the correct pronunciation and meaning of a word by using Google Translate. It uses advanced techniques of Natural Language Processing to achieve success in translating sentences into various languages.
2. Chatbots: To provide a better customer support service, companies have started using chatbots for 24/7 service. AI Chatbots help resolve the basic queries of customers. If a chatbot is not able to resolve any query, then it forwards it to the support team, while still engaging the customer. It helps make customers feel that the customer support team is quickly attending to them. With the help of chatbots, companies have become capable of building cordial relations with customers. It is only possible with the help of Natural Language Processing.

53. What are stop words?

Stop words are said to be useless data for a search engine. Words such as articles, prepositions, etc. are considered stop words. There are stop words such as was, were, is, am, the, a, an, how, why, and many more. In Natural Language Processing, we eliminate the stop words to understand and analyze the meaning of a sentence. The removal of stop words is one of the most important tasks for search engines. Engineers design the algorithms of search engines in such a way that they ignore the use of stop words. This helps show the relevant search result for a query.

54. What is NLTK?

NLTK is a Python library, which stands for Natural Language Toolkit. We use NLTK to process data in human-spoken languages. NLTK allows us to apply techniques such as parsing, tokenization, lemmatization, stemming, and more to understand natural languages. It helps in categorizing text, parsing linguistic structure, analyzing documents, etc.

A few of the libraries of the NLTK package that we often use in NLP are:

1. SequentialBackoffTagger
2. DefaultTagger
3. UnigramTagger
4. treebank
5. wordnet
6. FreqDist
7. patterns
8. RegexpTagger
9. backoff_tagger
10. UnigramTagger, BigramTagger, and TrigramTagger

55. What is Syntactic Analysis?

Syntactic analysis is a technique of analyzing sentences to extract meaning from them. Using syntactic analysis, a machine can analyze and understand the order of words arranged in a sentence. NLP employs grammar rules of a language that helps in the syntactic analysis of the combination and order of words in documents.

The techniques used for syntactic analysis are as follows:

1. Parsing: It helps in deciding the structure of a sentence or text in a document. It helps analyze the words in the text based on the grammar of the language.
2. Word segmentation: The segmentation of words segregates the text into small significant units.
3. Morphological segmentation: The purpose of morphological segmentation is to break words into their base form.
4. Stemming: It is the process of removing the suffix from a word to obtain its root word.
5. Lemmatization: It helps combine words using suffixes, without altering the meaning of the word.

56. What is Semantic Analysis?

Semantic analysis helps make a machine understand the meaning of a text. It uses various algorithms for the interpretation of words in sentences. It also helps understand the structure of a sentence.

Techniques used for semantic analysis are as given below:

1. Named entity recognition: This is the process of information retrieval that helps identify entities such as the name of a person, organization, place, time, emotion, etc.
2. Word sense disambiguation: It helps identify the sense of a word used in different sentences.
3. Natural language generation: It is a process used by the software to convert structured data into human-spoken languages. By using NLG, organizations can automate content for custom reports.

57. List the components of Natural Language Processing.

The major components of NLP are as follows:

• Entity extraction: Entity extraction refers to the retrieval of information such as place, person, organization, etc. by the segmentation of a sentence. It helps in the recognition of an entity in a text.
• Syntactic analysis: Syntactic analysis helps draw the specific meaning of a text.
• Pragmatic analysis: To find useful information from a text, we implement pragmatic analysis techniques.
• Morphological and lexical analysis: It helps in explaining the structure of words by analyzing them through parsing.

58. What is Latent Semantic Indexing (LSI)?

Latent semantic indexing is a mathematical technique used to improve the accuracy of the information retrieval process. The design of LSI algorithms allows machines to detect the hidden (latent) correlation between semantics (words). To enhance information understanding, machines generate various concepts that associate with the words of a sentence.

The technique used for information understanding is called singular value decomposition. It is generally used to handle static and unstructured data. The matrix obtained for singular value decomposition contains rows for words and columns for documents. This method is best suited to identify components and group them according to their types.

The main principle behind LSI is that words carry a similar meaning when used in a similar context. Computational LSI models are slow in comparison to other models. However, they are good at contextual awareness which helps improve the analysis and understanding of a text or a document.

59. What are Regular Expressions?

A regular expression is used to match and tag words. It consists of a series of characters for matching strings.

Suppose, if A and B are regular expressions, then the following are true for them:

• If {ɛ} is a regular language, then ɛ is a regular expression for it.
• If A and B are regular expressions, then A + B is also a regular expression within the language {A, B}.
• If A and B are regular expressions, then the concatenation of A and B (A.B) is a regular expression.
• If A is a regular expression, then A* (A occurring multiple times) is also a regular expression.

60. What is Regular Grammar?

Regular grammar is used to represent a regular language.

Regular grammar comprises rules in the form of A -> a, A -> aB, and many more. The rules help detect and analyze strings by automated computation.

Regular grammar consists of four tuples:

1. ‘N’ is used to represent the non-terminal set.
2. ‘∑’ represents the set of terminals.
3. ‘P’ stands for the set of productions.
4. ‘S € N’ denotes the start of non-terminal.

Regular grammar is of 2 types:
(a) Left Linear Grammar(LLG)

(b) Right Linear Grammar(RLG)

61. What is Parsing in the context of NLP?

Parsing in NLP refers to the understanding of a sentence and its grammatical structure by a machine. Parsing allows the machine to understand the meaning of a word in a sentence and the grouping of words, phrases, nouns, subjects, and objects in a sentence. Parsing helps analyze the text or the document to extract useful insights from it. To understand parsing, refer to the below diagram:

In this, ‘Jonas ate an orange’ is parsed to understand the structure of the sentence.

62. What is TF-IDF?

TFIDF or Term Frequency-Inverse Document Frequency indicates the importance of a word in a set. It helps in information retrieval with numerical statistics. For a specific document, TF-IDF shows a frequency that helps identify the keywords in a document. The major use of TF-IDF in NLP is the extraction of useful information from crucial documents by statistical data. It is ideally used to classify and summarize the text in documents and filter out stop words.

TF helps calculate the ratio of the frequency of a term in a document and the total number of terms. Whereas, IDF denotes the importance of the term in a document.

The formula for calculating TF-IDF:

TF(W) = (Frequency of W in a document)/(The total number of terms in the document)

IDF(W) = log_e(The total number of documents/The number of documents having the term W)

When TF*IDF is high, the frequency of the term is less and vice versa.

Google uses TF-IDF to decide the index of search results according to the relevancy of pages. The design of the TF-IDF algorithm helps optimize the search results in Google. It helps quality content rank up in search results.

63. Define the terminology in NLP.

The interpretation of Natural Language Processing depends on various factors, and they are:

Weights and Vectors

• Use of TF-IDF for information retrieval
• Length (TF-IDF and doc)
• Google Word Vectors
• Word Vectors

Structure of the Text

• POS tagging
• Head of the sentence
• Named Entity Recognition (NER)

Sentiment Analysis

• Knowledge of the characteristics of sentiment
• Knowledge about entities and the common dictionary available for sentiment analysis

Classification of Text

• Supervised learning algorithm
• Training set
• Validation set
• Test set
• Features of the text
• LDA

Machine Reading

• Removal of possible entities
• Joining with other entities
• DBpedia

FRED (lib) Pikes

64. Explain Dependency Parsing in NLP.

Dependency parsing helps assign a syntactic structure to a sentence. Therefore, it is also called syntactic parsing. Dependency parsing is one of the critical tasks in NLP. It allows the analysis of a sentence using parsing algorithms. Also, by using the parse tree in dependency parsing, we can check the grammar and analyze the semantic structure of a sentence.

For implementing dependency parsing, we use the spaCy package. It implements token properties to operate the dependency parse tree.

The below diagram shows the dependency parse tree:

65. What is the difference between NLP and NLU?

The below table shows the difference between NLP and NLU:

66. What is the difference between NLP and CI?

The below table shows the difference between NLP and CI:

67. What is Pragmatic Analysis?

Pragmatic analysis is an important task in NLP for interpreting knowledge that is lying outside a given document. The aim of implementing pragmatic analysis is to focus on exploring a different aspect of the document or text in a language. This requires a comprehensive knowledge of the real world. The pragmatic analysis allows software applications for the critical interpretation of the real-world data to know the actual meaning of sentences and words.

Example:

Consider this sentence: ‘Do you know what time it is?’

This sentence can either be asked for knowing the time or for yelling at someone to make them note the time. This depends on the context in which we use the sentence.

68. What is Pragmatic Ambiguity?

Pragmatic ambiguity refers to the multiple descriptions of a word or a sentence. An ambiguity arises when the meaning of the sentence is not clear. The words of the sentence may have different meanings. Therefore, in practical situations, it becomes a challenging task for a machine to understand the meaning of a sentence. This leads to pragmatic ambiguity.

Example:

Check out the below sentence.

‘Are you feeling hungry?’

The given sentence could be either a question or a formal way of offering food.

69. What are unigrams, bigrams, trigrams, and n-grams in NLP?

When we parse a sentence one word at a time, then it is called a unigram. The sentence parsed two words at a time is a bigram.

When the sentence is parsed three words at a time, then it is a trigram. Similarly, n-gram refers to the parsing of n words at a time.

Example: To understand unigrams, bigrams, and trigrams, you can refer to the below diagram:

Therefore, parsing allows machines to understand the individual meaning of a word in a sentence. Also, this type of parsing helps predict the next word and correct spelling errors.

70. What are the steps involved in solving an NLP problem?

Below are the steps involved in solving an NLP problem:

1. Gather the text from the available dataset or by web scraping
2. Apply stemming and lemmatization for text cleaning
3. Apply feature engineering techniques
4. Embed using word2vec
5. Train the built model using neural networks or other Machine Learning techniques
6. Evaluate the model’s performance
7. Make appropriate changes in the model
8. Deploy the model

71. What is Feature Extraction in NLP?

Features or characteristics of a word help in text or document analysis. They also help in sentiment analysis of a text. Feature extraction is one of the techniques that are used by recommendation systems. Reviews such as ‘excellent,’ ‘good,’ or ‘great’ for a movie are positive reviews, recognized by a recommender system. The recommender system also tries to identify the features of the text that help in describing the context of a word or a sentence. Then, it makes a group or category of the words that have some common characteristics. Now, whenever a new word arrives, the system categorizes it as per the labels of such groups.

72. What are precision and recall?

The metrics used to test an NLP model are precision, recall, and F1. Also, we use accuracy for evaluating the model’s performance. The ratio of prediction and the desired output yields the accuracy of the model.

Precision is the ratio of true positive instances and the total number of positively predicted instances.

Recall is the ratio of true positive instances and the total actual positive instances.

73. What is F1 score in NLP?

F1 score evaluates the weighted average of recall and precision. It considers both false negative and false positive instances while evaluating the model. F1 score is more accountable than accuracy for an NLP model when there is an uneven distribution of class. Let us look at the formula for calculating F1 score:

74. How to tokenize a sentence using the nltk package?

Tokenization is a process used in NLP to split a sentence into tokens. Sentence tokenization refers to splitting a text or paragraph into sentences.

For tokenizing, we will import sent_tokenize from the nltk package:

We will use the below paragraph for sentence tokenization:
Para = “Hi Guys. Welcome to Intellipaat. This is a blog on the NLP interview questions and answers.”

Output:

Tokenizing a word refers to splitting a sentence into words.

Now, to tokenize a word, we will import word_tokenize from the nltk package.

Para = “Hi Guys. Welcome to Intellipaat. This is a blog on the NLP interview questions and answers.”

Deep Learning

1. What is Deep Learning?

If you are going for a deep learning interview, you definitely know what exactly deep learning is. However, with this question the interviewee expects you to give an in-detail answer, with an example. Deep Learning involves taking large volumes of structured or unstructured data and using complex algorithms to train neural networks. It performs complex operations to extract hidden patterns and features (for instance, distinguishing the image of a cat from that of a dog).

 

 

2. What is a Neural Network?

Neural Networks replicate the way humans learn, inspired by how the neurons in our brains fire, only much simpler.

The most common Neural Networks consist of three network layers:

1. An input layer
2. A hidden layer (this is the most important layer where feature extraction takes place, and adjustments are made to train faster and function better)
3. An output layer

Each sheet contains neurons called “nodes,” performing various operations. Neural Networks are used in deep learning algorithms like CNN, RNN, GAN, etc.

3. What Is a Multi-layer Perceptron(MLP)?

As in Neural Networks, MLPs have an input layer, a hidden layer, and an output layer. It has the same structure as a single layer perceptron with one or more hidden layers. A single layer perceptron can classify only linear separable classes with binary output (0,1), but MLP can classify nonlinear classes.

Except for the input layer, each node in the other layers uses a nonlinear activation function. This means the input layers, the data coming in, and the activation function is based upon all nodes and weights being added together, producing the output. MLP uses a supervised learning method called “backpropagation.” In backpropagation, the neural network calculates the error with the help of cost function. It propagates this error backward from where it came (adjusts the weights to train the model more accurately).

4. What Is Data Normalization, and Why Do We Need It?

The process of standardizing and reforming data is called “Data Normalization.” It’s a pre-processing step to eliminate data redundancy. Often, data comes in, and you get the same information in different formats. In these cases, you should rescale values to fit into a particular range, achieving better convergence.

5. What is the Boltzmann Machine?

One of the most basic Deep Learning models is a Boltzmann Machine, resembling a simplified version of the Multi-Layer Perceptron. This model features a visible input layer and a hidden layer — just a two-layer neural net that makes stochastic decisions as to whether a neuron should be on or off. Nodes are connected across layers, but no two nodes of the same layer are connected.

6. What Is the Role of Activation Functions in a Neural Network?

At the most basic level, an activation function decides whether a neuron should be fired or not. It accepts the weighted sum of the inputs and bias as input to any activation function. Step function, Sigmoid, ReLU, Tanh, and Softmax are examples of activation functions.

 

 

7. What Is the Cost Function?

Also referred to as “loss” or “error,” cost function is a measure to evaluate how good your model’s performance is. It’s used to compute the error of the output layer during backpropagation. We push that error backward through the neural network and use that during the different training functions.

8. What Is Gradient Descent?

Gradient Descent is an optimal algorithm to minimize the cost function or to minimize an error. The aim is to find the local-global minima of a function. This determines the direction the model should take to reduce the error.

9. What Do You Understand by Backpropagation?

This is one of the most frequently asked deep learning interview questions. Backpropagation is a technique to improve the performance of the network. It backpropagates the error and updates the weights to reduce the error.

10. What Is the Difference Between a Feedforward Neural Network and Recurrent Neural Network?

In this deep learning interview question, the interviewee expects you to give a detailed answer.

A Feedforward Neural Network signals travel in one direction from input to output. There are no feedback loops; the network considers only the current input. It cannot memorize previous inputs (e.g., CNN).

A Recurrent Neural Network’s signals travel in both directions, creating a looped network. It considers the current input with the previously received inputs for generating the output of a layer and can memorize past data due to its internal memory.

11. What Are the Applications of a Recurrent Neural Network (RNN)?

The RNN can be used for sentiment analysis, text mining, and image captioning. Recurrent Neural Networks can also address time series problems such as predicting the prices of stocks in a month or quarter.

 

 

12. What Are the Softmax and ReLU Functions?

Softmax is an activation function that generates the output between zero and one. It divides each output, such that the total sum of the outputs is equal to one. Softmax is often used for output layers.

ReLU (or Rectified Linear Unit) is the most widely used activation function. It gives an output of X if X is positive and zeros otherwise. ReLU is often used for hidden layers.

13. What Are Hyperparameters?

This is another frequently asked deep learning interview question. With neural networks, you’re usually working with hyperparameters once the data is formatted correctly. A hyperparameter is a parameter whose value is set before the learning process begins. It determines how a network is trained and the structure of the network (such as the number of hidden units, the learning rate, epochs, etc.).

14. What Will Happen If the Learning Rate Is Set Too Low or Too High?

When your learning rate is too low, training of the model will progress very slowly as we are making minimal updates to the weights. It will take many updates before reaching the minimum point.

If the learning rate is set too high, this causes undesirable divergent behavior to the loss function due to drastic updates in weights. It may fail to converge (model can give a good output) or even diverge (data is too chaotic for the network to train).

15. What Is Dropout and Batch Normalization?

Dropout is a technique of dropping out hidden and visible units of a network randomly to prevent overfitting of data (typically dropping 20 percent of the nodes). It doubles the number of iterations needed to converge the network.

Batch normalization is the technique to improve the performance and stability of neural networks by normalizing the inputs in every layer so that they have mean output activation of zero and standard deviation of one.

The next step on this top Deep Learning interview questions and answers blog will be to discuss intermediate questions.

16. What Is the Difference Between Batch Gradient Descent and Stochastic Gradient Descent?

Batch Gradient Descent	Stochastic Gradient Descent
The batch gradient computes the gradient using the entire dataset. It takes time to converge because the volume of data is huge, and weights update slowly.	The stochastic gradient computes the gradient using a single sample. It converges much faster than the batch gradient because it updates weight more frequently.

 

 

17. What is Overfitting and Underfitting, and How to Combat Them?

Overfitting occurs when the model learns the details and noise in the training data to the degree that it adversely impacts the execution of the model on new information. It is more likely to occur with nonlinear models that have more flexibility when learning a target function. An example would be if a model is looking at cars and trucks, but only recognizes trucks that have a specific box shape. It might not be able to notice a flatbed truck because there’s only a particular kind of truck it saw in training. The model performs well on training data, but not in the real world.

Underfitting alludes to a model that is neither well-trained on data nor can generalize to new information. This usually happens when there is less and incorrect data to train a model. Underfitting has both poor performance and accuracy.

To combat overfitting and underfitting, you can resample the data to estimate the model accuracy (k-fold cross-validation) and by having a validation dataset to evaluate the model.

18. How Are Weights Initialized in a Network?

There are two methods here: we can either initialize the weights to zero or assign them randomly.

Initializing all weights to 0: This makes your model similar to a linear model. All the neurons and every layer perform the same operation, giving the same output and making the deep net useless.

Initializing all weights randomly: Here, the weights are assigned randomly by initializing them very close to 0. It gives better accuracy to the model since every neuron performs different computations. This is the most commonly used method.

19. What Are the Different Layers on CNN?

There are four layers in CNN:

1. Convolutional Layer –  the layer that performs a convolutional operation, creating several smaller picture windows to go over the data.
2. ReLU Layer – it brings non-linearity to the network and converts all the negative pixels to zero. The output is a rectified feature map.
3. Pooling Layer – pooling is a down-sampling operation that reduces the dimensionality of the feature map.
4. Fully Connected Layer – this layer recognizes and classifies the objects in the image.

Master the deep learning concepts and TensorFlow open-source framework with the Deep Learning Course with TensorFlow Certification. Check out the course preview today.

20. What is Pooling on CNN, and How Does It Work?

Pooling is used to reduce the spatial dimensions of a CNN. It performs down-sampling operations to reduce the dimensionality and creates a pooled feature map by sliding a filter matrix over the input matrix.

 

 

21. How Does an LSTM Network Work?

Long-Short-Term Memory (LSTM) is a special kind of recurrent neural network capable of learning long-term dependencies, remembering information for long periods as its default behavior. There are three steps in an LSTM network:

• Step 1: The network decides what to forget and what to remember.
• Step 2: It selectively updates cell state values.
• Step 3: The network decides what part of the current state makes it to the output.

22. What Are Vanishing and Exploding Gradients?

While training an RNN, your slope can become either too small or too large; this makes the training difficult. When the slope is too small, the problem is known as a “Vanishing Gradient.” When the slope tends to grow exponentially instead of decaying, it’s referred to as an “Exploding Gradient.” Gradient problems lead to long training times, poor performance, and low accuracy.

 

 

23. What Is the Difference Between Epoch, Batch, and Iteration in Deep Learning?

• Epoch – Represents one iteration over the entire dataset (everything put into the training model).
• Batch – Refers to when we cannot pass the entire dataset into the neural network at once, so we divide the dataset into several batches.
• Iteration – if we have 10,000 images as data and a batch size of 200. then an epoch should run 50 iterations (10,000 divided by 50).

24. Why is Tensorflow the Most Preferred Library in Deep Learning?

Tensorflow provides both C++ and Python APIs, making it easier to work on and has a faster compilation time compared to other Deep Learning libraries like Keras and Torch. Tensorflow supports both CPU and GPU computing devices.

25. What Do You Mean by Tensor in Tensorflow?

This is another most frequently asked deep learning interview question. A tensor is a mathematical object represented as arrays of higher dimensions. These arrays of data with different dimensions and ranks fed as input to the neural network are called “Tensors.”

26. What Are the Programming Elements in Tensorflow?

Constants – Constants are parameters whose value does not change. To define a constant we use  tf.constant() command. For example:

a = tf.constant(2.0,tf.float32)

b = tf.constant(3.0)

Print(a, b)

Variables – Variables allow us to add new trainable parameters to graph. To define a variable, we use the tf.Variable() command and initialize them before running the graph in a session. An example:

W = tf.Variable([.3].dtype=tf.float32)

b = tf.Variable([-.3].dtype=tf.float32)

Placeholders – these allow us to feed data to a tensorflow model from outside a model. It permits a value to be assigned later. To define a placeholder, we use the tf.placeholder() command. An example:

a = tf.placeholder (tf.float32)

b = a*2

with tf.Session() as sess:

result = sess.run(b,feed_dict={a:3.0})

print result

Sessions – a session is run to evaluate the nodes. This is called the “Tensorflow runtime.” For example:

a = tf.constant(2.0)

b = tf.constant(4.0)

c = a+b

# Launch Session

Sess = tf.Session()

# Evaluate the tensor c

print(sess.run(c))

27. Explain a Computational Graph.

Everything in a tensorflow is based on creating a computational graph. It has a network of nodes where each node operates, Nodes represent mathematical operations, and edges represent tensors. Since data flows in the form of a graph, it is also called a “DataFlow Graph.”

28. Explain Generative Adversarial Network.

Suppose there is a wine shop purchasing wine from dealers, which they resell later. But some dealers sell fake wine. In this case, the shop owner should be able to distinguish between fake and authentic wine.

The forger will try different techniques to sell fake wine and make sure specific techniques go past the shop owner’s check. The shop owner would probably get some feedback from wine experts that some of the wine is not original. The owner would have to improve how he determines whether a wine is fake or authentic.

The forger’s goal is to create wines that are indistinguishable from the authentic ones while the shop owner intends to tell if the wine is real or not accurately.

Let us understand this example with the help of an image shown above.

There is a noise vector coming into the forger who is generating fake wine.

Here the forger acts as a Generator.

The shop owner acts as a Discriminator.

The Discriminator gets two inputs; one is the fake wine, while the other is the real authentic wine. The shop owner has to figure out whether it is real or fake.

So, there are two primary components of Generative Adversarial Network (GAN) named:

1. Generator
2. Discriminator

The generator is a CNN that keeps keys producing images and is closer in appearance to the real images while the discriminator tries to determine the difference between real and fake images The ultimate aim is to make the discriminator learn to identify real and fake images.

29. What Is an Auto-encoder?

This Neural Network has three layers in which the input neurons are equal to the output neurons. The network’s target outside is the same as the input. It uses dimensionality reduction to restructure the input. It works by compressing the image input to a latent space representation then reconstructing the output from this representation.

 

 

30. What Is Bagging and Boosting?

Bagging and Boosting are ensemble techniques to train multiple models using the same learning algorithm and then taking a call.

With Bagging, we take a dataset and split it into training data and test data. Then we randomly select data to place into the bags and train the model separately.

With Boosting, the emphasis is on selecting data points which give wrong output to improve the accuracy.

The following are some of the most important advanced deep learning interview questions that you should know!

31. What is the significance of using the Fourier transform in Deep Learning tasks?

The Fourier transform function efficiently analyzes, maintains, and manages large datasets. You can use it to generate real-time array data that is helpful for processing multiple signals.

32. What do you understand by transfer learning? Name a few commonly used transfer learning models.

Transfer learning is the process of transferring the learning from a model to another model without having to train it from scratch. It takes critical parts of a pre-trained model and applies them to solve new but similar machine learning problems.

Some of the popular transfer learning models are:

• VGG-16
• BERT
• GTP-3
• Inception V3
• XCeption

33. What is the difference between SAME and VALID padding in Tensorflow?

Using the Tensorflow library, tf.nn.max_pool performs the max-pooling operation. Tf.nn.max_pool has a padding argument that takes 2 values – SAME or VALID.

With padding == “SAME” ensures that the filter is applied to all the elements of the input.

The input image gets fully covered by the filter and specified stride. The padding type is named SAME as the output size is the same as the input size (when stride=1).

With padding == “VALID” implies there is no padding in the input image. The filter window always stays inside the input image. It assumes that all the dimensions are valid so that the input image gets fully covered by a filter and the stride defined by you.

34. What are some of the uses of Autoencoders in Deep Learning?

• Autoencoders are used to convert black and white images into colored images.
• Autoencoder helps to extract features and hidden patterns in the data.
• It is also used to reduce the dimensionality of data.
• It can also be used to remove noises from images.

35. What is the Swish Function?

Swish is an activation function proposed by Google which is an alternative to the ReLU activation function.

It is represented as: f(x) = x * sigmoid(x).

The Swish function works better than ReLU for a variety of deeper models.

The derivative of Swist can be written as: y’ = y + sigmoid(x) * (1 – y)

36. What are the reasons for mini-batch gradient being so useful?

• Mini-batch gradient is highly efficient compared to stochastic gradient descent.
• It lets you attain generalization by finding the flat minima.
• Mini-batch gradient helps avoid local minima to allow gradient approximation for the whole dataset.

37. What do you understand by Leaky ReLU activation function?

Leaky ReLU is an advanced version of the ReLU activation function. In general, the ReLU function defines the gradient to be 0 when all the values of inputs are less than zero. This deactivates the neurons. To overcome this problem, Leaky ReLU activation functions are used. It has a very small slope for negative values instead of a flat slope.

38. What is Data Augmentation in Deep Learning?

Data Augmentation is the process of creating new data by enhancing the size and quality of training datasets to ensure better models can be built using them. There are different techniques to augment data such as numerical data augmentation, image augmentation, GAN-based augmentation, and text augmentation.

39. Explain the Adam optimization algorithm.

Adaptive Moment Estimation or Adam optimization is an extension to the stochastic gradient descent. This algorithm is useful when working with complex problems involving vast amounts of data or parameters. It needs less memory and is efficient.

Adam optimization algorithm is a combination of two gradient descent methodologies –

Momentum and Root Mean Square Propagation.

40. Why is a convolutional neural network preferred over a dense neural network for an image classification task?

• The number of parameters in a convolutional neural network is much more diminutive than that of a Dense Neural Network. Hence, a CNN is less likely to overfit.
• CNN allows you to look at the weights of a filter and visualize what the network learned. So, this gives a better understanding of the model.
• CNN trains models in a hierarchical way, i.e., it learns the patterns by explaining complex patterns using simpler ones.

41. Which strategy does not prevent a model from over-fitting to the training data?

1. Dropout
2. Pooling
3. Data augmentation
4. Early stopping

Answer: b) Pooling – It’s a layer in CNN that performs a downsampling operation.

42. Explain two ways to deal with the vanishing gradient problem in a deep neural network.

• Use the ReLU activation function instead of the sigmoid function
• Initialize neural networks using Xavier initialization that works with tanh activation.

43. Why is a deep neural network better than a shallow neural network?

Both deep and shallow neural networks can approximate the values of a function. But the deep neural network is more efficient as it learns something new in every layer. A shallow neural network has only one hidden layer. But a deep neural network has several hidden layers that create a deeper representation and computation capability.

 

 

44. What is the need to add randomness in the weight initialization process?

If you set the weights to zero, then every neuron at each layer will produce the same result and the same gradient value during backpropagation. So, the neural network won’t be able to learn the function as there is no asymmetry between the neurons. Hence, randomness to the weight initialization process is crucial.

45. How can you train hyperparameters in a neural network?

Hyperparameters in a neural network can be trained using four components:

Batch size: Indicates the size of the input data.

Epochs: Denotes the number of times the training data is visible to the neural network to train.

Momentum: Used to get an idea of the next steps that occur with the data being executed.

Learning rate: Represents the time required for the network to update the parameters and learn

Computer Vision

 

Generative AI

1. What is Generative AI?

Generative AI, short for Generative Artificial Intelligence, is a subset of artificial intelligence (AI) that focuses on enabling machines to produce content or data that resembles human-generated information. It’s a technology that’s gaining immense popularity in various fields, from natural language processing to creative content generation.

Generative AI operates on a principle of learning patterns from existing data and using that knowledge to create new content. It relies on deep learning techniques, particularly neural networks, to accomplish this task. These neural networks are trained on large datasets, allowing them to generate text, images, music, and more.

2. How does Generative AI work?

Generative AI works through the use of neural networks, specifically Recurrent Neural Networks (RNNs) and more recently, Transformers. Here’s a simplified breakdown of how it functions:

• Data Collection: To begin, a substantial amount of data related to the specific task is gathered. For instance, if you want to generate text, the model needs a massive text corpus to learn from.
• Training: The neural network is then trained on this data. During training, the model learns the underlying patterns, structures, and relationships within the data. It learns to predict the next word, character, or element in a sequence.
• Generation: Once trained, the model can generate content by taking a seed input and predicting the subsequent elements. For instance, if you give it the start of a sentence, it can complete the sentence in a coherent and contextually relevant manner.
• Fine-Tuning: Generative AI models can be further fine-tuned for specific tasks or domains to improve the quality of generated content.

3. What are the top applications of Generative AI?

Generative AI has a wide range of applications across different industries:

• Natural Language Processing (NLP): It’s used for text generation, language translation, and chatbots that can engage in human-like conversations.
• Content Generation: Generative AI can create articles, stories, and even poetry. It’s used by content creators to assist in writing.
• Image and Video Generation: It can generate realistic images and videos, which are valuable in fields like entertainment and design.
• Data Augmentation: In data science, it’s used to create synthetic data for training machine learning models.
• Healthcare: Generative AI helps in generating medical reports, simulating disease progression, and drug discovery.

4. Can you explain the difference between Generative AI and Discriminative AI?

Aspect	Generative AI	Discriminative AI
Objective	Generates new data based on learned patterns.	Classifies input data into predefined classes.
Function	Content creation, data generation.	Classification, discrimination.
Use Cases	Text generation, image synthesis, creativity.	Spam detection, sentiment analysis, image recognition.
Learning Process	Learns patterns in data for content generation.	Focuses on learning boundaries between classes.
Examples	Chatbots, text generators, art creation.	Spam filters, image classifiers, sentiment analysis models.

5. What are some popular Generative AI models?

Generative AI models have revolutionized the field of artificial intelligence, offering remarkable capabilities in generating content, from text to images and beyond. In this section, we’ll explore some of the most popular and influential Generative AI models that have left a significant mark on the industry.

1. a. GPT-4 (Generative Pre-trained Transformer 4): GPT-4, developed by OpenAI, is a standout among Generative AI models. With billions of parameters, it has demonstrated remarkable text generation abilities. GPT-4 can answer questions, write essays, generate code, and even create conversational agents that engage users in natural language.
2. BERT (Bidirectional Encoder Representations from Transformers): Although primarily known for its prowess in natural language understanding, BERT also exhibits generative capabilities. It excels in tasks like text completion and summarization, making it a valuable tool in various applications, including search engines and chatbots.
3. DALL·E: If you’re interested in generative art, DALL·E is a model to watch. Developed by OpenAI, this model can generate images from textual descriptions. It takes creativity to new heights by creating visuals based on written prompts, showing the potential of Generative AI in the visual arts.
4. StyleGAN2: When it comes to generating realistic images, StyleGAN2 is a name that stands out. It can create high-quality, diverse images that are virtually indistinguishable from real photographs. StyleGAN2 has applications in gaming, design, and even fashion.
5. VQ-VAE-2 (Vector Quantized Variational Autoencoder 2): This model combines elements of generative and variational autoencoders to generate high-quality, high-resolution images. It has made significant strides in image compression and generation.

6. How is Generative Adversarial Networks (GANs) used in AI?

Generative Adversarial Networks, or GANs, have emerged as a groundbreaking concept in the realm of Generative AI. These networks consist of two primary components: a generator and a discriminator, which work in tandem to create and evaluate content. Here’s how GANs are used in AI:

Application	Description
Image Generation	Create high-quality images, artworks, and more.
Data Augmentation	Generate additional data to enhance training datasets.
Style Transfer	Transform the style of images, e.g., artist-inspired styles.
Super-Resolution	Enhance image resolution for clarity and detail.
Anomaly Detection	Identify deviations from the normal data distribution.
Text-to-Image Generation	Generate images from textual descriptions.

Top Applications of GANs:

• Image Generation: GANs are widely used to create high-quality images, such as faces of non-existent individuals, realistic artworks, and more. This is particularly valuable in creative industries and design.
• Data Augmentation: GANs can generate additional data to augment training datasets. This is crucial in scenarios where obtaining large amounts of real data is challenging.
• Style Transfer: GANs can transform the style of images, such as converting a photograph into the style of a famous artist. This has applications in art and design.
• Super-Resolution: GANs can enhance the resolution of images, making them sharper and clearer. This is beneficial in fields like medical imaging and photography.
• Anomaly Detection: GANs can be used to detect anomalies in data by learning the normal distribution of data and flagging deviations from it.
• Text-to-Image Generation: GANs can generate images from textual descriptions, opening up possibilities in e-commerce and visual storytelling.

7. What are the limitations of Generative AI?

While Generative AI has made remarkable strides, it’s essential to acknowledge its limitations and challenges. Understanding these limitations is crucial for responsible and effective use. Here are some key constraints of Generative AI:

1. Data Dependency: Generative AI models, including GANs, require vast amounts of data for training. Without sufficient data, the quality of generated content may suffer, and the model might produce unrealistic or biased results.
2. Ethical Concerns: Generative AI can inadvertently perpetuate biases present in the training data. This raises ethical concerns, particularly when it comes to generating content related to sensitive topics, such as race, gender, or religion.
3. Lack of Control: Generative AI can be unpredictable. Controlling the output to meet specific criteria, especially in creative tasks, can be challenging. This lack of control can limit its practicality in some applications.
4. Resource Intensive: Training and running advanced Generative AI models demand substantial computational resources, making them inaccessible to smaller organizations or individuals with limited computing power.
5. Overfitting: Generative models may memorize the training data instead of learning its underlying patterns. This can result in content that lacks diversity and creativity.
6. Security Risks: There is the potential for malicious use of Generative AI, such as generating deepfake videos for deceptive purposes or creating fake content to spread misinformation.
7. Intellectual Property Concerns: When Generative AI is used to create content, determining ownership and copyright becomes complex. This raises legal questions about intellectual property rights.
8. Validation Challenges: It can be difficult to validate the authenticity of content generated by Generative AI, which can be problematic in contexts where trust and reliability are paramount.

8. What are the ethical concerns surrounding Generative AI?

Generative AI, with its ability to create content autonomously, brings forth a host of ethical considerations. As this technology becomes more powerful, it’s crucial to address these concerns to ensure responsible and ethical use. Here are some of the ethical concerns surrounding Generative AI:

1. Bias and Fairness: Generative AI models can inadvertently perpetuate biases present in their training data. This can lead to the generation of content that reflects and reinforces societal biases related to race, gender, and other sensitive attributes.
2. Privacy: Generative AI can be used to create deepfake content, including fabricated images and videos that can infringe upon an individual’s privacy and reputation.
3. Misinformation: The ease with which Generative AI can generate realistic-looking text and media raises concerns about its potential for spreading misinformation and fake news.
4. Identity Theft: Generative AI can create forged identities, making it a potential tool for identity theft and fraud.
5. Deceptive Content: Malicious actors can use Generative AI to create deceptive content, such as fake reviews, emails, or social media posts, with the intent to deceive or defraud.
6. Legal and Copyright Issues: Determining the legal ownership and copyright of content generated by AI can be complex, leading to legal disputes and challenges.
7. Psychological Impact: The use of Generative AI in creating content for entertainment or social interactions may have psychological impacts on individuals who may not always distinguish between AI-generated and human-generated content.
8. Accountability: Establishing accountability for content generated by AI is challenging. When harmful content is created, it can be unclear who should be held responsible.

To address these ethical concerns, developers and users of Generative AI must prioritize responsible and ethical practices. This includes rigorous data curation to minimize bias, clear labeling of AI-generated content, and adherence to ethical guidelines and regulations.

9. How can Generative AI be used in art and creativity?

Use Case	Description
Art Generation	AI algorithms create visual art based on input parameters.
Music Composition	AI generates music, offering fresh inspiration to musicians.
Writing Assistance	AI assists writers with ideas, plot twists, and even stories.
Design Optimization	AI optimizes layouts, colors, and styles in design fields.
Art Restoration	AI reconstructs damaged artworks, preserving cultural heritage.
Style Transfer	AI applies artistic styles to photos or images, creating unique visuals.
Virtual Worlds	AI powers immersive virtual worlds, enhancing gaming and entertainment.
Fashion Design	AI generates clothing designs, predicting trends in fashion.

10. What are the challenges in training Generative AI models?

Training Generative AI models presents several challenges:

1. Data Quality: High-quality training data is essential. Noisy or biased data can lead to flawed outputs.
2. Computational Resources: Training large models demands substantial computational power and time.
3. Mode Collapse: GANs may suffer from mode collapse, where they generate limited varieties of outputs.
4. Ethical Considerations: AI-generated content can raise ethical issues, including misinformation and deepfakes.
5. Evaluation Metrics: Measuring the quality of generated content is subjective and requires robust evaluation metrics.

11. What are the key components of a GAN architecture?

A Generative Adversarial Network (GAN) comprises two main components:

1. Generator: This component creates synthetic data. It takes random noise as input and transforms it into data that resembles the training dataset.
2. Discriminator: The discriminator’s role is to distinguish between real and generated data. It learns to classify data as real or fake.

GANs operate on a feedback loop. The generator aims to produce data that can fool the discriminator, while the discriminator gets better at distinguishing real from fake data. This competition results in the generation of high-quality synthetic content.

12. How does text generation with Generative AI work?

Text generation with Generative AI involves models like GPT (Generative Pre-trained Transformer). Here’s how it works:

1. Pre-training: Models are initially trained on a massive corpus of text data, learning grammar, context, and language nuances.
2. Fine-tuning: After pre-training, models are fine-tuned on specific tasks or datasets, making them domain-specific.
3. Autoregressive Generation: GPT generates text autoregressively, predicting the next word based on context. It’s conditioned on input text.
4. Sampling Strategies: Techniques like beam search or temperature-based sampling control the creativity and diversity of generated text.

13. Can Generative AI create realistic images and videos?

Generative AI, including models like Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs), has made remarkable strides in creating realistic images and videos. These technologies are at the forefront of modern artificial intelligence, bridging the gap between creativity and technology.

Generative AI accomplishes this feat by learning from vast datasets of real-world images and videos. It then employs a two-step process to generate new content. Here’s how it works:

• Generator Network: The generator network takes random noise as input and attempts to produce data that resembles real images or videos. This network is responsible for the creative aspect, introducing variations and uniqueness into the content.
• Discriminator Network: Simultaneously, there’s a discriminator network that evaluates the content generated by the generator. Its role is to distinguish between real and generated content. It’s like a detective trying to spot fake art from genuine masterpieces.

These two networks engage in a continuous battle. The generator aims to produce content that fools the discriminator into believing it’s real, while the discriminator becomes increasingly skilled at telling the difference. This back-and-forth training process eventually results in the generator creating highly realistic images and videos.

14. How does StyleGAN work, and what are its applications?

StyleGAN is a cutting-edge Generative Adversarial Network (GAN) variant renowned for its ability to generate high-resolution, realistic images with an unprecedented level of control and customization.

At its core, StyleGAN operates by separating the generation process into two crucial components: the style and the structure.

• Style Mapping: StyleGAN starts by mapping a latent vector (essentially a set of random numbers) into a style space. This style space controls various high-level features of the generated image, such as the pose, facial expression, and overall aesthetics. This separation of style from structure allows for precise control over these attributes.
• Synthesis Network: The second part involves a synthesis network that generates the image structure based on the learned style. This network uses convolutional layers to create the image pixel by pixel, guided by the style information. This separation of style and structure allows for incredible flexibility and customization.

Applications:

Applications of StyleGAN	Description
Art and Fashion	Create customizable art pieces and fashion designs with unique aesthetics.
Facial Generation	Generate realistic faces for video games, digital characters, and movie special effects.
Data Augmentation	Diversify datasets for machine learning, improving model training and performance.
Content Creation	Produce unique visuals, logos, and branding materials for various creative purposes.
Realistic Image Editing	Edit images while maintaining authenticity, enabling advanced image manipulation.

15. Are there any Generative AI models used in natural language processing (NLP)?

Generative AI models have made significant strides in the field of Natural Language Processing (NLP), revolutionizing the way machines understand and generate human language. One of the most prominent examples is the use of Transformers, a class of generative models that has reshaped NLP.

Transformers, which includes models like GPT-4 (Generative Pre-trained Transformer 4) and BERT (Bidirectional Encoder Representations from Transformers), have demonstrated remarkable capabilities in understanding and generating natural language text.

Here’s how they work:

• Attention Mechanism: Transformers utilize an attention mechanism that allows them to weigh the importance of each word or token in a sentence concerning others. This mechanism helps the model capture context effectively.
• Pre-training: These models are pre-trained on a vast corpora of text data. During this phase, they learn grammar, facts, and even some reasoning abilities from the text. For example, they can predict the next word in a sentence or mask a word and predict it based on the surrounding context.
• Fine-tuning: After pre-training, models like GPT-3 or BERT are fine-tuned on specific NLP tasks like language translation, sentiment analysis, or question-answering. This fine-tuning tailors the model to excel in these particular tasks.

16. How can Generative AI be used in healthcare?

Generative AI in Healthcare	Technical Applications
Medical Imaging	Enhancing image quality for diagnosis.
Drug Discovery	Generating molecular structures for new drugs.
Health Data Generation	Synthesizing medical data for ML datasets.
Predictive Modeling	Creating models for disease outbreak prediction.
Natural Language Processing	Generating medical reports and clinical notes.
Personalized Medicine	Tailoring treatment plans based on patient data.
Medical Simulations	Creating realistic training simulations for healthcare professionals.

17. What role does reinforcement learning play in Generative AI?

Reinforcement learning, a pivotal branch of artificial intelligence, plays a substantial role in the realm of Generative AI. At its core, reinforcement learning involves training models to make sequences of decisions by interacting with an environment.

In Generative AI, reinforcement learning is often employed to enhance the generation process.

Here’s how it works: 

The AI model generates an output, such as an image or text; then, reinforcement learning comes into play by evaluating the quality of that output. If it’s subpar, the model adjusts its internal parameters to generate better results.

This iterative process continues, gradually improving the AI’s ability to create content. It’s particularly beneficial when precision and fine-tuning are essential, as in applications like natural language generation and image synthesis.

18. What is the importance of data in training Generative AI models?

Data is the lifeblood of Generative AI models. The quality and quantity of data used in training have a profound impact on the model’s performance. Generative AI models learn from data, seeking patterns and structures within it to generate new content.

For instance, in text generation, a model trained on a diverse and extensive dataset can produce more coherent and contextually relevant text. In image generation, the richness of data influences the model’s ability to create high-resolution and visually pleasing images.

Moreover, data diversity is vital. Training data should encompass various styles, contexts, and nuances to enable the AI model to adapt to different scenarios. Without robust data, Generative AI models would lack the foundation needed for creativity and accuracy.

19. Can Generative AI be used for anomaly detection?

Yes, Generative AI can be a powerful tool for anomaly detection. Anomaly detection involves identifying patterns or instances that deviate significantly from the norm within a dataset. Generative AI models, such as autoencoders and GANs (Generative Adversarial Networks), excel in this area.

Autoencoders, for example, are neural networks designed to reconstruct their input data. When trained on normal data, they become adept at reproducing it accurately. However, when presented with anomalies, they struggle to reconstruct them accurately, highlighting deviations.

Similarly, GANs can generate data that mimics the training dataset’s characteristics. Any data that significantly differs from the generated samples is flagged as an anomaly. This application is valuable in various domains, including fraud detection and cybersecurity.

20. What are some examples of Generative AI generating music?

Generative AI Music Tools	Key Features
Meta’s AudioCraft	– Trained on licensed music and sound effects. – Enables quick addition of music and sounds to videos via text prompts.
OpenAI’s MuseNet	– Analyzes style, rhythm, and harmony in music. – Can switch between music genres and incorporate up to 10 instruments.
iZotope’s AI Assistants	– Pioneering AI-assisted music production tool. – Offers valuable insights and assistance in music creation.
Jukebox by OpenAI	– Generates music samples from scratch based on genre, artist, and lyrics.
VEED’s AI Music Generator	– Creates royalty-free, unique soundtracks for videos using Generative AI.

21. How does Generative AI impact content generation on the internet?

Aspect	Description
Efficiency	Rapidly generates large amounts of content
Personalization	Tailors content to individual user preferences
Multilingual Support	Creates content in multiple languages
SEO Optimization	Analyzes keywords for better search engine ranking
Content Variability	Produces diverse content types for wider engagement
Quality Control	Requires human oversight for accuracy and coherence

22. What are some successful real-world applications of Generative AI?

Application	Example
Image Generation	OpenAI’s DALL-E generated an image from text descriptions
Conversational AI Apps for Patients	Ada: Symptom assessment and medical guidance in multiple languages
AI for Early Disease Detection	SkinVision: Early detection of skin cancer
AI for Accessibility	Be My Eyes: Converting images to text for the visually impaired
AI for Patient Interactions and Support	Hyro: Enhancing patient engagement and healthcare support
Content Creation	ChatGPT: Generating text content and creative writing

23. How do you evaluate the quality of output from a Generative AI model?

Evaluation Aspect	Description
Human Review	Assess output for coherence, relevance, and accuracy
Diversity Check	Ensure content doesn’t become repetitive
Plagiarism Detection	Verify originality and copyright compliance
User Feedback	Gather user input for improvement
Domain-Specific Metrics	Use metrics like BLEU scores for specific domains
Ethical Considerations	Ensure content aligns with ethical guidelines

24. Can Generative AI be used for language translation?

Yes, Generative AI is increasingly used for language translation, and it has significantly improved the accuracy and efficiency of translation services. Here’s how it works:

• Neural Machine Translation (NMT): Generative AI models, particularly those based on NMT, excel at language translation. They analyze vast amounts of bilingual text data to learn how languages correspond and then generate translations based on this knowledge.
• Multilingual Capabilities: These models can handle multiple languages, making them versatile for global communication.
• Continuous Improvement: AI translation models continuously learn and adapt to language nuances, ensuring that translations become more accurate over time.
• Real-time Translation: AI-powered translation services are integrated into various platforms, allowing for real-time translation of text, speech, and even images.

25. What are the privacy concerns related to Generative AI?

Privacy concerns surrounding Generative AI have become increasingly prominent in recent years. As these powerful AI models, like GPT-4, continue to evolve, several key issues have emerged:

• Data Privacy: Generative AI models require vast amounts of data to train effectively. This raises concerns about the privacy of the data used, as it may include sensitive or personal information.
• Bias and Fairness: Generative AI models can inadvertently perpetuate biases present in their training data. This can lead to biased or unfair outputs, impacting various applications from content generation to decision-making.
• Deepfakes and Misinformation: Generative AI can be used to create highly convincing deepfake videos and text, making it challenging to distinguish between real and fabricated content, thus fueling the spread of misinformation.
• Security Risks: Malicious actors can misuse Generative AI to automate phishing attacks, create fake identities, or generate fraudulent content, posing significant security risks.
• User Privacy: As AI models generate personalized content, there is a concern about user privacy. How much personal information should be input for customization, and how securely is it stored?

To address these concerns, researchers and developers are actively working on improving transparency, fairness, and privacy-preserving techniques in Generative AI. It’s crucial to strike a balance between the power of these models and the potential risks they pose to privacy.

26. How can Generative AI models be fine-tuned for specific tasks?

Steps	Description
Step 1: Dataset Selection	Choose a relevant, diverse dataset.
Step 2: Architecture Selection	Pick a suitable pre-trained model.
Step 3: Task-Specific Objective	Define a clear task and adapt the model.
Step 4: Hyperparameter Tuning	Adjust parameters for optimal performance.
Step 5: Training Process	Train the model and monitor performance.
Step 6: Regularization Techniques	Apply techniques like dropout and decay.
Step 7: Evaluation	Assess performance using relevant metrics.

27. What are some challenges in making Generative AI models more efficient?

Efficiency is a critical aspect of Generative AI models. Several challenges need to be overcome to make these models more efficient:

• Computational Resources: Training and running large AI models demands significant computational power, making them inaccessible for many users.
• Model Size: The sheer size of models like GPT-3 poses challenges in terms of memory and storage requirements.
• Inference Speed: Real-time applications require models that can generate responses quickly, which can be a challenge for complex Generative AI models.
• Energy Consumption: Running large models consumes a substantial amount of energy, which is not environmentally sustainable.
• Scalability: Scaling up AI models to handle diverse tasks while maintaining efficiency is a complex task.

28. Can Generative AI be used for generating 3D models?

Yes, Generative AI can be harnessed for 3D model generation. This exciting application has gained traction in recent years. Here’s how it works:

• Data Preparation: Generative AI models require 3D training data, which can include images, point clouds, or even existing 3D models.
• Model Architecture: Specialized architectures like 3D-GANs (Generative Adversarial Networks) or VAEs (Variational Autoencoders) are used for 3D model generation.
• Training: The model is trained to generate 3D structures based on the provided data. This can be used for creating 3D objects, scenes, or even medical images.
• Applications: 3D Generative AI finds applications in various fields, including gaming, architectural design, medical imaging, and manufacturing, enabling the automated creation of 3D content.

29. How does Generative AI assist in generating new product designs?

Generative AI is revolutionizing the field of product design. It leverages deep learning algorithms to analyze vast datasets of existing designs, user preferences, and market trends. By doing so, it assists designers in generating innovative and unique product concepts. Here’s how it works:

Generative AI algorithms, such as Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs), learn patterns and features from large datasets of product designs.

These algorithms can then generate new design variations based on the learned patterns. This not only accelerates the design process but also opens the door to entirely novel ideas.

Designers can input specific constraints or preferences, and Generative AI will adapt the generated designs accordingly. This level of customization is a game-changer in product development.

Generative AI also aids in rapid prototyping, allowing designers to explore multiple design options quickly.

In summary, Generative AI empowers designers by offering a wealth of design possibilities, streamlining the creative process, and ultimately leading to the creation of more innovative products.

30. Are there any Generative AI models that generate code?

Yes, there are Generative AI models specifically designed for code generation. These models are a boon for developers, as they automate and optimize many aspects of software development. Here’s an overview:

• One prominent example is OpenAI’s GPT-4, which can generate code snippets for a variety of programming languages.
• Another noteworthy model is OpenAI’s Codex, built on GPT-3, which excels at understanding and generating code in natural language. It’s like having a coding companion.
• Generative AI models analyze code repositories and documentation to understand coding conventions and best practices. They can then generate code that aligns with these standards.
• These models are not just limited to generating simple code snippets; they can assist in more complex tasks, such as writing entire functions or even suggesting optimized algorithms.
• Developers can save time and reduce errors by leveraging Generative AI models for code generation, making software development more efficient.

31. What is the role of Generative AI in generating realistic game environments?

Generative AI plays a pivotal role in the gaming industry, enhancing the creation of immersive and realistic game environments. Here’s how it contributes:

• Generative AI algorithms, particularly procedural content generation (PCG), can generate vast and diverse game worlds. These algorithms use mathematical rules to create terrain, landscapes, and structures, reducing the need for manual design.
• Realistic textures and 3D models can be generated with the help of Generative AI, making game environments visually stunning.
• Dynamic storytelling within games benefits from Generative AI’s ability to create branching narratives and adapt to player choices, resulting in a more engaging player experience.
• Generative AI can simulate natural behaviors for in-game characters, making NPCs (non-playable characters) and enemies more lifelike and responsive.

32. How does Generative AI influence storytelling and narrative generation?

Generative AI is reshaping storytelling by providing powerful tools for authors and content creators. Here’s how it influences narrative generation:

• Generative AI models can analyze vast amounts of text data to understand storytelling patterns, character development, and plot structures.
• Authors can use Generative AI to brainstorm ideas, generate plot outlines, or even create character dialogues that fit seamlessly within a story.
• These models can adapt their output to match a specific writing style or genre, making them versatile tools for authors with varying creative needs.
• In collaborative writing, Generative AI can suggest plot twists, character arcs, or even entire chapters, fostering creativity and efficiency among writers.
• Generative AI is not here to replace authors but to assist and inspire, making the storytelling process more efficient and imaginative.

33. Can Generative AI be used for data augmentation in machine learning?

Generative AI, a remarkable branch of artificial intelligence, plays a pivotal role in enhancing machine learning models through data augmentation. It’s a technique that resonates with both beginners and seasoned professionals.

Data augmentation is the process of increasing the diversity and volume of training data to improve the robustness and accuracy of machine learning models. Generative AI, with its ability to generate synthetic data, has found a crucial application in this domain.

Using Generative Adversarial Networks (GANs) and other generative techniques, data scientists can create realistic data points that closely mimic the distribution of the original dataset. This synthetic data can then be added to the training set, effectively increasing its size and variety.

The benefits are twofold. First, it helps prevent overfitting by providing more examples for the model to learn from. Second, it aids in addressing data scarcity issues, especially in niche domains where collecting extensive data is challenging.

However, it’s essential to ensure that the generated data is of high quality and representative of the real-world scenarios. Rigorous validation and testing are crucial steps in this process to maintain the integrity of the model.

34. What are the future prospects of Generative AI?

Aspect	Future Outlook
Improved Realism	Enhanced realism in generated content.
Industry Applications	Diverse applications across industries.
Ethical Considerations	Addressing ethical concerns around deepfakes.
Healthcare Innovations	Advancements in medical imaging and drug discovery.
Personalized Content	Tailored content generation for users in real-time.
Climate Modeling	Improved climate change simulations and predictions.

35. How is Generative AI used in generative design in architecture and engineering?

Generative AI revolutionizes architecture and engineering by enabling generative design. This approach leverages algorithms to explore countless design possibilities. Here’s how Generative AI is making its mark:

Generative algorithms analyze parameters and constraints provided by architects and engineers to generate designs. This iterative process leads to innovative solutions, optimizing structures for functionality, aesthetics, and sustainability.

36. What are the security implications of Generative AI?

Concern	Description
Deepfakes	Realistic fake videos for potential misuse.
Data Privacy	Privacy risks related to data generation.
Authentication	Challenges to traditional authentication methods.
Identity Theft	Potential for generating convincing fake identities.
Content Manipulation	Altered content may deceive and spread misinformation.
Cyberattacks	AI-generated cyber threats and attacks.

37. How can Generative AI be used in personalized content recommendation?

Generative AI plays a pivotal role in personalized content recommendation systems. By utilizing advanced machine learning algorithms, Generative AI tailors content suggestions based on individual preferences and behavior. Here’s how it works:

Generative AI models, such as GPT-4, analyze user data like past browsing history, search queries, and interactions with content. These models then generate recommendations that are highly relevant to each user.

For example, if you’re an e-commerce platform, Generative AI can suggest products based on a user’s previous purchases and the preferences of similar users. This personalization enhances user engagement and boosts conversion rates.

38. What are the hardware requirements for training large Generative AI models?

Hardware Component	Requirements and Description
Graphics Processing Units	High-performance GPUs or TPUs are essential for processing complex computations during model training. Multiple GPUs in a cluster can significantly speed up the process.
Memory Capacity	Large memory capacity is crucial for storing model parameters, especially in the case of large Generative AI models.
Storage	Fast storage solutions, such as Solid State Drives (SSDs), are used to enable quick data retrieval and storage during training.
Computing Clusters	Distributed computing clusters with multiple GPUs are employed for parallel processing, reducing training time.
Internet Connection	Access to high-speed internet is necessary for downloading and transferring large datasets, as well as for accessing cloud-based resources for training.

39. How does unsupervised learning relate to Generative AI?

Unsupervised learning is at the core of Generative AI. It’s a machine learning paradigm where models learn from unlabeled data, finding hidden patterns and structures. Generative AI leverages unsupervised learning to create data or content that resembles human-generated data.

For example, unsupervised learning is used in Generative Adversarial Networks (GANs), a popular Generative AI architecture. GANs consist of a generator and a discriminator network that compete with each other. The generator aims to create realistic data (like images or text), while the discriminator tries to distinguish between real and generated data. This adversarial process drives the generator to produce increasingly authentic content.

40. Can Generative AI be used in drug discovery and molecular design?

Absolutely, Generative AI is a game-changer in drug discovery and molecular design. It expedites the process of identifying potential drug candidates and designing new molecules with specific properties.

Generative AI models can predict molecular properties, generate novel chemical structures, and optimize existing compounds. This accelerates drug development, making it more cost-effective and efficient.

Researchers and pharmaceutical companies are utilizing Generative AI to simulate molecular interactions, screen potential drugs, and discover novel solutions for challenging medical conditions.

41. What are the common evaluation metrics used to gauge the performance of Large Language Models (LLMs) in Generative AI?

Evaluation metrics encompass perplexity, BLEU score, ROUGE score, METEOR score, F1 score, and human evaluation criteria. These metrics assess various facets such as coherence, relevance, diversity, and fluency of the generated text, ensuring comprehensive evaluation of LLMs in Generative AI.

Database

 

1. What is SQL?

SQL means Structured Query Language and is used to communicate with relational databases. It proposes a standardized way to interact with databases, allowing users to perform various operations on the data, including retrieval, insertion, updating, and deletion.

2. What are the different types of SQL commands?

• SELECT: Retrieves data from a database.
• INSERT: Adds new records to a table.
• UPDATE: Modifies existing records in a table.
• DELETE: Removes records from a table.
• CREATE: Creates a new database, table, or view.
• ALTER: Modifies the existing database object structure.
• DROP: Deletes an existing database object.

3. What is a primary key in SQL?

It is a unique identifier for each record in a table. It ensures that each row in the table has a distinct and non-null value in the primary key column. Primary keys enforce data integrity and create relationships between tables.

4. What is a foreign key?

It is a field in one table referencing the primary key in another. It establishes a relationship between the two tables, ensuring data consistency and enabling data retrieval across tables.

5. Explain the difference between DELETE and TRUNCATE commands.

The DELETE command is used by professionals to remove particular rows from a table based on a condition, allowing you to selectively delete records. TRUNCATE, on the other hand, removes all rows from a table without specifying conditions. TRUNCATE is faster and uses fewer system resources than DELETE but does not log individual row deletions.

6. What is a JOIN in SQL, and what are its types?

A JOIN operation merges information from two or more tables by utilizing a common column that links them together. Various types of JOINs exist, like INNER JOIN, LEFT JOIN, RIGHT JOIN, and FULL JOIN. These JOIN variations dictate the manner in which data from the involved tables is paired and retrieved.

7. What do you mean by a NULL value in SQL?

A NULL value in SQL represents the absence of data in a column. It is not the same as an empty string or zero; it signifies that the data is missing or unknown. NULL values can be used in columns with optional data or when the actual data is unavailable.

8. Define a Unique Key in SQL.

Often referred to as a unique constraint, a unique key guarantees that every value in a column (or a combination of columns) remains distinct and cannot be replicated within a table. In contrast to a primary key, a table has the flexibility to incorporate multiple unique keys.

9. What is a database?

A database is a systematically organized collection of data arranged into tables composed of rows and columns. The primary purpose of databases is to efficiently store, manage, and retrieve data.

10. Explain the differences between SQL and NoSQL databases.

SQL databases are characterized by their use of structured tables and strict adherence to a predefined schema, making them ideal for managing structured data with a strong focus on data consistency and transaction support. In contrast, NoSQL databases are non-relational and excel in handling unstructured or semi-structured data, frequently employed for scalable, distributed, and adaptable data storage solutions.

11. What is a table and a field in SQL?

In SQL, a table is a structured data collection organized into rows and columns. Each column in a table is called a field, representing a specific attribute or property of the data.

12. Describe the SELECT statement.

The SELECT statement serves the purpose of fetching data from one or multiple tables, enabling you to specify the desired columns to retrieve, apply filters through the WHERE clause, and manage the result’s sorting using the ORDER BY clause.

13. What is a constraint in SQL? Name a few.

A constraint in SQL defines rules or restrictions that apply to data in a table, ensuring data integrity. Common constraints include:

• PRIMARY KEY: Ensures the values’ uniqueness in a column.
• FOREIGN KEY: Enforces referential integrity between tables.
• UNIQUE: Ensures the uniqueness of values in a column.
• CHECK: Defines a condition that data must meet to be inserted or updated.
• NOT NULL: Ensures that there are no NULL values in a column.

14. What is normalization in SQL?

Normalization is the method used to streamline data storage within a database, reducing redundancy and enhancing data integrity. This approach entails dividing tables into more manageable, interrelated tables and establishing connections between them.

15. How do you use the WHERE clause?

The WHERE clause within SQL queries serves the purpose of selectively filtering rows according to specified conditions, thereby enabling you to fetch exclusively those rows that align with the criteria you define. For example:

SELECT * FROM employees WHERE department = ‘HR’;

Become a Software Development Professional

• 

  Full Stack Java Developer

  • Kickstart Full Stack Java Developer career with industry-aligned curriculum by experts
  • Hands-on practice through 20+ projects, assessments, and tests

  6 months

  View Program
• 

  Full Stack Web Developer – MEAN Stack

  • Comprehensive Blended Learning program
  • 8X higher interaction in live online classes conducted by industry experts

  11 months

  View Program

Not sure what you’re looking for?View all Related Programs

16. What are indexes in SQL?

Indexes improve the data retrieval operations speed. They provide a quick way to locate specific rows in a table by creating a sorted data structure based on one or more columns. Indexes are essential for optimizing query performance.

17. Explain GROUP BY in SQL.

The GROUP BY clause organizes rows from a table into groups based on the values in one or more columns. It is commonly employed alongside aggregate functions like SUM, COUNT, AVG, MIN, and MAX to perform computations on data that has been grouped together.

18. What is an SQL alias?

An SQL alias serves as a transitory label bestowed upon either a table or a column within a query, with the primary purpose of enhancing the clarity of query outcomes or simplifying the process of renaming columns for improved referencing. For example:

SELECT first_name AS “First Name”, last_name AS “Last Name” FROM employees;

19. Explain ORDER BY in SQL.

The ORDER BY clause is used to sort the result set of a query based on one or more columns. You can specify each column’s sorting order (ascending or descending). For example:

SELECT * FROM products ORDER BY price DESC;

20. Describe the difference between WHERE and HAVING in SQL.

The WHERE clause is employed to restrict individual rows before they are grouped, such as when filtering rows prior to a GROUP BY operation. Conversely, the HAVING clause is utilized to filter groups of rows after they have been grouped, like filtering groups based on aggregate values.

21. What is a view in SQL?

An SQL view is essentially a virtual table that derives its data from the outcome of a SELECT query. Views serve multiple purposes, including simplifying intricate queries, enhancing data security through an added layer, and enabling the presentation of targeted data subsets to users, all while keeping the underlying table structure hidden.

22. What is a stored procedure?

A SQL stored procedure comprises precompiled SQL statements that can be executed together as a unified entity. These procedures are commonly used to encapsulate business logic, improve performance, and ensure consistent data manipulation practices.

23. What is a trigger in SQL?

An SQL trigger consists of a predefined sequence of actions that are executed automatically when a particular event occurs, such as when an INSERT or DELETE operation is performed on a table. Triggers are employed to ensure data consistency, conduct auditing, and streamline various tasks.

 

24. What are aggregate functions? Can you name a few?

Aggregate functions in SQL perform calculations on a set of values and return a single result.

• SUM: To calculate the sum of values in a column.
• COUNT: To count a column’s number of rows or non-null values.
• AVG: To calculate the average of values in a column.
• MIN: To retrieve the minimum value in a column.
• MAX: To retrieve the maximum value in a column.

25. How do you update a value in SQL?

The UPDATE statement serves the purpose of altering pre-existing records within a table. It involves specifying the target table for the update, the specific columns to be modified, and the desired new values to be applied. For example:

UPDATE employees SET salary = 60000 WHERE department = ‘IT’;

Intermediate SQL Interview Questions and Answers

26. What is a self-join, and how would you use it?

A self-join is a type of join where a table is joined with itself. It is useful when creating relationships within the same table, such as finding hierarchical relationships or comparing rows with related data.

27. Explain different types of joins with examples.

• INNER JOIN: Gathers rows that have matching values in both tables.
• RIGHT JOIN: Gathers all rows from the right table and any matching rows from the left table.
• LEFT JOIN: Gathers all rows from the left table and any matching rows from the right table.
• FULL JOIN: Gathers all rows when there’s a match in either table, including unmatched rows from both tables.

Example:

SELECT employees.name, departments.name

FROM employees

LEFT JOIN departments ON employees.department_id = departments.id;

28. What is a subquery? Provide an example.

A subquery refers to a query that is embedded within another query, serving the purpose of fetching information that will subsequently be employed as a condition or value within the encompassing outer query. For example, to find employees with salaries greater than the average salary:

SELECT name

FROM employees

WHERE salary > (SELECT AVG(salary) FROM employees);

29. How do you optimize SQL queries?

SQL query optimization involves improving the performance of SQL queries by reducing resource usage and execution time. Strategies include using appropriate indexes, optimizing query structure, and avoiding costly operations like full table scans.

 

Become a Software Development Professional

• 

  Full Stack Java Developer

  • Kickstart Full Stack Java Developer career with industry-aligned curriculum by experts
  • Hands-on practice through 20+ projects, assessments, and tests

  6 months

  View Program
• 

  Full Stack Web Developer – MEAN Stack

  • Comprehensive Blended Learning program
  • 8X higher interaction in live online classes conducted by industry experts

  11 months

  View Program

Not sure what you’re looking for?View all Related Programs

 

30. What is the difference between UNION and UNION ALL?

UNION merges the outcomes of two or more SELECT statements, removing duplicate rows, whereas UNION ALL merges the results without removing duplicates. While UNION ALL is faster, it may include duplicate rows.

31. What are correlated subqueries?

It is a type of subquery that makes reference to columns from the surrounding outer query. This subquery is executed repeatedly, once for each row being processed by the outer query, and its execution depends on the outcomes of the outer query.

32. Explain ACID properties in SQL.

ACID – Atomicity, Consistency, Isolation, and Durability. They are essential properties that ensure the reliability and integrity of database transactions:

• Atomicity (single, indivisible unit of transactions)
• Consistency (transactions bring the DB from one consistent state to another)
• Isolation (transactions are isolated from each other)
• Durability (committed transactions are permanent and survive system failures)

33. What is a transaction in SQL?

A transaction in SQL is a sequence of one or more SQL operations treated as a single unit of work. Transactions ensure that database operations are either completed successfully or rolled back entirely in case of failure.

34. How do you implement error handling in SQL?

Error handling in SQL is typically achieved using try-catch blocks (in SQL Server) or EXCEPTION blocks (in Oracle). These blocks allow you to handle and log errors gracefully to prevent application crashes.

35. What is a cursor, and how is it used?

In SQL, a cursor is a database element employed for the purpose of fetching and controlling data one row at a time from a result set. Cursors find frequent application within stored procedures or triggers when it becomes necessary to process data in a sequential manner.

36. Describe the data types in SQL.

SQL supports various data types, including numeric, character, date/time, and binary types. Common data types include INT, VARCHAR, DATE, and BLOB, among others. Data types define the kind of values a column can hold.

37. Explain normalization and denormalization.

Normalization is the method used to streamline data in a database, decreasing redundancy and enhancing data integrity. This procedure includes dividing large tables into smaller, interconnected ones to eliminate duplicated data. Conversely, denormalization is the deliberate act of introducing redundancy to enhance query performance.

38. What is a clustered index?

A clustered index in SQL determines the physical order of data rows in a table. Each table can have only one clustered index, which impacts the table’s storage structure. Rows in a table are physically stored in the same order as the clustered index key.

39. How do you prevent SQL injection?

SQL injection represents a security flaw that arises when SQL queries mishandle untrusted data, posing a risk of unauthorized access or data tampering. To ward off SQL injection, employ techniques like parameterized queries, prepared statements, input validation, and the enforcement of stringent access controls.

40. What are the different types of triggers?

• DML triggers: These triggers fire in response to data manipulation language (DML) operations such as INSERT, UPDATE, or DELETE.
• DDL triggers: These triggers fire in response to data definition language (DDL) events, such as table or view creation.

41. Explain the concept of a database schema.

In SQL, a database schema functions as a conceptual container for housing various database elements, such as tables, views, indexes, and procedures. Its primary purpose is to facilitate the organization and segregation of these database elements while specifying their structure and interconnections.

42. How is data integrity ensured in SQL?

Data integrity in SQL is ensured through various means, including constraints (e.g., primary keys, foreign keys, check constraints), normalization, transactions, and referential integrity constraints. These mechanisms prevent invalid or inconsistent data from being stored in the database.

43. What is an SQL injection?

SQL injection is a cybersecurity attack method that involves the insertion of malicious SQL code into an application’s input fields or parameters. This unauthorized action enables attackers to illicitly access a database, extract confidential information, or manipulate data.

44. How do you create a stored procedure?

You use the CREATE PROCEDURE statement to create a stored procedure in SQL. A stored procedure can contain SQL statements, parameters, and variables. Here’s a simple example:

CREATE PROCEDURE GetEmployeeByID(@EmployeeID INT)

AS

BEGIN

SELECT * FROM employees WHERE employee_id = @EmployeeID;

END;

45. What is a deadlock in SQL? How can it be prevented?

A deadlock in SQL occurs when two or more transactions cannot proceed because they are waiting for resources held by each other. Deadlocks can be prevented or resolved by using techniques such as locking hierarchies, timeouts, or deadlock detection and resolution mechanisms.

 

Advanced SQL Interview Questions

46. Explain different isolation levels in SQL.

Isolation levels define the visibility of data changes one transaction makes to other concurrent transactions. There are four commonly used isolation levels in SQL:

• READ UNCOMMITTED: At this isolation level, transactions are allowed to read changes made by other transactions even if those changes have not been committed. While this provides the highest level of concurrency, it also introduces the risk of encountering dirty reads.
• READ COMMITTED: In this level, transactions can only read committed data, avoiding dirty reads. However, it may still suffer from non-repeatable reads and phantom reads.
• REPEATABLE READ: Transactions at this level ensure that any data read during the transaction remains unchanged throughout the transaction’s lifetime. It prevents non-repeatable reads but may still allow phantom reads.
• SERIALIZABLE: This represents the utmost isolation level, guaranteeing absolute isolation between transactions. While it eradicates all concurrency problems, it may exhibit reduced efficiency due to locking mechanisms.

47. How does a clustered index work and how is it different from a non-clustered index?

A clustered index defines the actual storage order of rows within a table, allowing for only one clustered index per table and directly influencing the on-disk data organization. Conversely, a non-clustered index does not impact the physical arrangement of data and can coexist with multiple indexes within the same table.

• Clustered Index: When you create a clustered index on a table, the table’s rows are physically rearranged to match the order of the indexed column(s). This makes range queries efficient but may slow down insert/update operations.
• Non-clustered Index: Non-clustered indexes are separate data structures that store a copy of a portion of the table’s data and point to the actual data rows. They improve read performance but come with some overhead during data modification.

48. Discuss SQL server reporting services.

SQL Server Reporting Services is a reporting tool provided by Microsoft for creating, managing, and delivering interactive, tabular, graphical, and free-form reports. SSRS allows users to design and generate reports from various data sources, making it a valuable asset for businesses needing comprehensive reporting capabilities.

49. What are ctes (Common table expressions)?

Common Table Expressions (CTEs) serve as momentary result sets that you can mention within SQL statements, typically found within SELECT, INSERT, UPDATE, or DELETE operations. They’re established using the `WITH` keyword and are instrumental in streamlining intricate queries by dividing them into more digestible components.

50. Explain the MERGE statement.

The SQL MERGE statement is employed to execute insertions, updates, or deletions on a target table, guided by the outcomes of a source table or query. It consolidates the functionalities of several individual statements (INSERT, UPDATE, DELETE) into one comprehensive statement, rendering it particularly valuable for achieving data synchronization between tables.

51. How do you use a window function in SQL?

Window functions are employed to carry out computations on a group of table rows that are associated with the current row. They enable the generation of result sets containing aggregated data while retaining the distinct details of each row. Typical window functions encompass ROW_NUMBER(), RANK(), DENSE_RANK(), and SUM() OVER().

52. What is a pivot table and how do you create one in SQL?

A pivot table is a technique used to rotate or transpose rows into columns to better analyze and summarize data. You can create pivot tables in SQL using the `PIVOT` operator to convert row-based data into a column-based format.

53. Describe the process of database mirroring.

Database mirroring is a high-availability solution in SQL Server that involves creating and maintaining redundant copies of a database on separate servers. It ensures data availability and disaster recovery by automatically failing over to the mirror server in case of a primary server failure.

54. Explain the concept of table partitioning.

Partitioning a table involves the strategy of breaking down a sizable table into smaller, more easily handled segments known as partitions. This method can enhance query efficiency by permitting SQL Server to focus solely on pertinent partitions while executing queries. Typically, partitioning is carried out using a column characterized by a high cardinality, such as date or region.

55. How do you handle transactions in distributed databases?

Handling transactions in distributed databases involves ensuring the ACID (Atomicity, Consistency, Isolation, Durability) properties across multiple databases or nodes. This can be achieved through distributed transaction management protocols like Two-Phase Commit (2PC) or by using distributed database systems designed for this purpose.

56. What is the use of the explain plan?

The EXPLAIN plan is a valuable feature found in numerous relational database management systems. This tool offers a comprehensive view of the database engine’s strategy for executing a query, encompassing details such as the selected execution plan, join techniques, index utilization, and projected costs. Database administrators (DBAs) and developers rely on EXPLAIN plans to enhance the performance of their queries.

57. Discuss SQL server integration services (SSIS).

Microsoft provides SQL Server Integration Services as a powerful ETL (Extract, Transform, Load) tool. It enables data integration from various sources, transformation of data as needed, and loading it into destination systems like data warehouses or databases.

58. What are indexed views?

Indexed views, or materialized views, are precomputed result sets stored as physical tables in the database. They improve query performance by allowing the database engine to access pre-aggregated or pre-joined data directly from the indexed view, reducing the need for complex query processing.

Become a Software Development Professional

• 

  Full Stack Java Developer

  • Kickstart Full Stack Java Developer career with industry-aligned curriculum by experts
  • Hands-on practice through 20+ projects, assessments, and tests

  6 months

  View Program
• 

  Full Stack Web Developer – MEAN Stack

  • Comprehensive Blended Learning program
  • 8X higher interaction in live online classes conducted by industry experts

  11 months

  View Program

Not sure what you’re looking for?View all Related Programs

59. Explain the concept of database sharding.

Database sharding is a horizontal partitioning technique that distributes data across multiple database instances or servers. It’s commonly used in large-scale systems to improve scalability and performance. Each shard contains a subset of the data, and a sharding strategy determines how data is distributed.

60. How do you manage large-scale databases for performance?

Managing large-scale databases for performance involves various strategies, including proper indexing, partitioning, query optimization, hardware optimization, and caching. Monitoring and fine-tuning the database is crucial to ensure optimal performance as data volumes grow.

61. What is a materialized view?

Materialized views are a type of database component designed to maintain the outcomes of a query in the form of a tangible table. These views undergo periodic updates to ensure that the stored data remains current. They are employed to enhance the efficiency of database queries, particularly for intricate or frequently executed ones.

62. Discuss the strategies for database backup and recovery.

Ensuring data availability and disaster recovery relies on the implementation of vital backup and recovery strategies. These strategies encompass various methods, such as full backups, differential backups, transaction log backups, and regular testing of restoration procedures.

63. What are the best practices for securing a SQL database?

Securing a SQL database involves implementing access controls, encryption, auditing, and regular security assessments. Best practices include using strong authentication, limiting permissions, and keeping database systems and software up to date.

64. Explain the concept of database replication.

It is the process of copying and synchronizing data from one database to another. It ensures data availability, load balancing, and disaster recovery. Common replication types include snapshot replication, transactional replication, and merge replication.

65. How do you monitor SQL server performance?

Monitoring SQL Server performance involves tracking key performance metrics, setting up alerts for critical events, and analyzing performance bottlenecks. Tools like SQL Server Profiler and Performance Monitor are commonly used for this purpose.

66. What is a database warehouse?

A database warehouse is a centralized repository that stores data from various sources for analytical and reporting purposes. It is optimized for querying and analysis and often contains historical data.

67. Explain the use of full-text search in SQL.

Full-text search in SQL allows users to search for text-based data within large text fields or documents. It uses advanced indexing and search algorithms to provide efficient and accurate text-searching capabilities.

68. How do you manage database concurrency?

Database concurrency involves a database system’s capability to manage multiple concurrent transactions while upholding data integrity. To achieve this, various techniques such as locking mechanisms, optimistic concurrency control, and isolation levels are employed to oversee and regulate database concurrency.

69. What are the challenges in handling big data in SQL?

Handling big data in SQL involves dealing with large volumes of data that exceed the capabilities of traditional database systems. Challenges include data storage, processing, scalability, and efficient querying. Solutions may include distributed databases and big data technologies like Hadoop and Spark.

70. How do you implement high availability in SQL databases?

High availability in SQL databases ensures that the database remains accessible and operational despite failures. Techniques like clustering, replication, and failover mechanisms help achieve high availability.

71. Explain the use of XML data type in SQL server.

The XML data type allows to store, retrieve, and manipulate XML data. It provides support for querying XML documents using XQuery, and it’s commonly used in applications that deal with XML data structures.

72. Discuss the concept of NoSQL databases and their interaction with SQL.

NoSQL databases are non-relational databases designed for handling large volumes of unstructured or semi-structured data. They interact with SQL databases through various integration methods, such as data pipelines, ETL processes, and API-based data transfers.

73. What is a spatial database?

A spatial database stores and queries geometric and geographic data, such as maps, GPS coordinates, and spatial objects. It provides specialized functions and indexing methods to support spatial queries and analysis.

74. How do you migrate a database from one server to another?

Database migration involves moving a database from one server or platform to another. This undertaking demands careful planning, data transfer, schema conversion, and thorough testing to ensure a smooth transition while minimizing the potential for data loss or system downtime.

75. Discuss advanced optimization techniques for SQL queries.

Advanced optimization techniques for SQL queries include using query hints, indexing strategies, query rewriting, and understanding the query execution plan. Profiling tools and performance monitoring are essential for identifying and resolving performance bottlenecks.

API

1. What is REST?

REST stands for Representational State Transfer.

2. What is a REST API?

An API is an application programming interface, which is a software-to-software interface that allows otherwise separate applications to interact and share data. In a REST API, all data is treated as resources, each one represented by a unique uniform resource identifier (URI).

3. What do you mean by RESTful web services?

REST API is also known as RESTful web services that follow the REST architecture.

4. What are cache-control headers?

Cache-control headers are used to control catching and to attain caching ability. The most commonly used cache-control headers are public, private, and No-Store.

5. What are the features of RESTful web services?

REStful web services have the following unique features:

• Client-server decoupling
• Communication support
• Lightweight
• Uniform interface
• Stateless
• Layered system
• Cacheable
• Code on demand

6. What is the definition of messaging in terms of RESTful web services?

In REST API web services, when a REST client wants to send a message to the server, it can be sent in an HTTP request form, and the same applies to the server. This kind of communication is called messaging in REST.

7. Explain ‘Addressing’ in RESTful web services.

The process of locating various types of resources with the help of a URL on the REST server is known as ‘addressing’ in RESTful web services. Usually, single or multiple resources are addressed by resources.

8. Why are REST services easily scalable?

REST services are scalable due to the statelessness that they do not store data on the server even though they are requested and do not require much communication.

9. What are Idempotent methods?

Idempotent methods are known to return the same outcome even after the same request has been made multiple times, and it avoids errors caused by duplicate requests on the client side.

10. How can RESTful web services be tested?

The RESTful web services can be tested with the help of tools such as Swagger and Postman, which enable users to inspect query parameters, response headers, and headers, documentation of the endpoints, and conversion of endpoints to XML and JSON.

11. What are payloads in RESTful web services?

Payloads are the request data passed through the POST or GET method and found in the message’s body of an HTTP request in RESTful web services.

12. What is the maximum payload size that can be sent in POST methods?

Theoretically, there is no such maximum limit for payload size that can be sent in POST methods. However, payloads with larger sizes can consume larger bandwidth. Thus the server could take more time to proceed with the request.

13. Which protocol does REST APIs use?

Protocols are used to communicate with clients where REST APIs use HTTP protocol for it.

14. In REST APIs, which markup languages are used to represent the resources?

The resources in REST APIs are represented with the help of XML (extensible markup language) and JSON (JavaScript Object Notation).

15. Differentiate POST and PUT methods.

POST Method

• POST can create a resource on the server.
• POST is not idempotent.
• POST responses are cacheable.

PUT Method

• PUT is used to replace a resource at a specific URI with another resource.
• PUT is idempotent that it will only result in one resource even after calling it multiple times.
• PUT responses are not.

16. Which HTTP request methods are supported by REST?

REST supports various types of HTTP request methods such as GET, POST, PUT, DELETE, HEAD, OPTIONS, ETC.

17. What is CRUD?

CRUD stands for “Create, Read, Update, Delete.”

18. The main parts of an HTTP response

The main parts of the HTTP response are the HTTP version, Status line, HTTP Response Header, and HTTP Response body.

19. What are the most common HTTP response status codes you see while working in REST API?

Some of the most common response status codes are 200 OK, 201 Created, 400 Bad Request, 401 Unauthorized, 403 Forbidden, 404 Not Found, 500 Internal Server Error, 502 Bad Gateway, 503 Service Unavailable, etc.

20. What is a resource?

In REST, A resource is an object with a label and accessible on the server. Resources consist of associated data, a list of methods, and a relationship with other resources on the server.

21. What is a URI?

URI stands for ‘Uniform Resource Identifier.

22. What is caching in the REST API?

REST API stores a copy of a server response in a particular location of computer memory to retrieve the server response fast in the future. This method is temporary and called “catching.”

23. What’s a real-world example of a REST API?

1. Public REST APIs are harnessed by weather apps to display weather information and share the related data.
2. Airlines use APIs to expose the flight times and prices to allow travel and ticketing sites for businesses.
3. Public transportation services use APIs to make their data publicly open to make it available for mapping and navigation apps in real-time.

24. What is the difference between REST and SOAP?

REST(Representational State Transfer)

• It is an architectural design pattern used to develop web services.
• It is faster in speed and more cacheable.
• It inherits only the security measures concerning the protocol that have been implemented.

SOAP (Simple Object Access Protocol)

• It is a strict protocol used to build secure APIs.
• It is slower in speed and not cacheable.
• It is able to define its own security measures.

25. What do you understand about JAX-RS?

It is a Java-based specification implemented for RESTful services and defined by JEE.

26. Disadvantages of RESTful web services?

• RESTful web services are stateless and do not maintain session simulation responsibility as the client side does not provide a particular session id for it.
• REST is not able to impose the security restriction inherently. However, it inherits them with the help of implementing protocols. Thus, the integration of SSL/TLS authentication needs to be done very carefully for better security measures of the REST APIs.

27. Advantages of REST

• HTTP makes the implementation of REST easy.
• REST fits in the existing infrastructure of the web, thus the web application can easily implement the REST. XML and JSON web technologies make REST easy to learn.
• The client and server communication is stateless, thus the integration is easy to build and scalable, and manageable with respect to time.
• The REST architecture can adapt to a huge variety of cases due to its flexibility.
• The lightweight architecture of REST makes it easy to build the applications faster as compared to other types of APIs.
• REST can be tested easily in the browser with the help of API testing tools.

28. How do you keep REST APIs secure?

REST APIs can be kept secure with the help of safety measures such as Authentication and authorization, API Server Validation, TSl/SSL Encryption, Rate-limiting for DDoS attacks, and sensitive information such as username, password, or authentication token should not be visible in URIs

29. What are “Options” in REST APIs?

It is an HTTP method used to fetch the supported HTTP options or operations that help clients to choose the options in REST APIs. Cross-Origin Resource Sharing (CORS) uses the REST option method.

30. Different types of API architectures

There are other two API architectures used, SOAP (Simple Object Access Protocol), and RPC (Remote Procedure Call)

31. What are the different application integration styles?

The different application integration styles are Shared database, Batch file transfer, Invoking remote procedure (RPC), and Swapping asynchronous messages over a message-oriented middleware (MOM).

32. How is JAXB related to RESTful web API?

JAXB is a Java arch used for XML binding in RESTful web API.

33. What is AJAX?

AJAX stands for  Asynchronous javascript and XML.

34. What does the HEAD method in REST APIs do?

The HEAD method is used to return the HTTP Header in read-only form and not the Body.

35. Which frameworks can JAX-RS implement in the RESTful web?

JAX-RS is used to implement frameworks such as Jersey, RESTEasy, Apache, and CFX.

36. What are HTTP status codes and their meaning?

• Code 200: success.
• Code 201:resource has been successfully created.
• Code 204: no content in the response body.
• Code 404: no method available.

37. What is a ‘Resource’?

‘Resource’ is defined as an object of a type that includes image, HTML file, text data, and any type of dynamic data.

38. Why is the proper representation of resources required?

Proper representations of resources in the proper format allow the client to easily understand the format and determine the identification of resources easily.

39. How to design Resources representation for RESTful web services?

• It should be easy to understand for the client and server.
• It should be complete irrespective of its format structure.
• It should consider the link of the resources to other resources and handle it carefully.

40. Important aspects of RESTful web services implementation.

• ResourcesRequest
• Headers
• Request Body
• Response Body
• Status codes

Flask

1. What do you know about Flask?

Ans: Flask is a Python-based framework that builds web applications. The interface is mostly based on HTTP, REST, GraphQL, or Websockets. This framework is based on the Jinja2 templates engine and WSGI web application library. The creator, Armin Ronacher, developed it for the Pallets project initially.

If you want to enrich your career and become a professional in Python, then enroll in “Python Online Training” – This course will help you to achieve excellence in this domain.

2. Is Flask an open-source framework?

Ans: Yes, Flask is an open-source framework.

3. Why do we use the Flask framework in web application development?

Ans: Flask framework is based on the Python programming language. It has a quick microframework based on prototyping web and networking applications to execute the code faster. We can use Flask in the below-mentioned cases:

1. When we need to develop an API or ML Model.
2. To control security camera by API abstraction.
3. When we need to work on no-SQL like DynamoDB.
4. When we need to do ElasticSearch.
5. When we need to prepare a microservice adapter to translate SOAP API into JSON.

4. How can we download the Flask development version?

Ans: We can get the Flask development version by using the given below command:

git clone https://github.com/pallets/flask
cd flask && python3 setup.py develop

5. Do you know how to install Flask in Linux?

Ans: You can install Flask in a Linux environment using Python package manager, pip.

Related Article: Python Tutorial for Beginners

6. Can we add an emailing function in Flask?

Yes, we can add an emailing function in the Flask framework. Actually, we need to install the Flask-Mail extension. We can use the given below command:

Pip install Flask-Mail 

Now, after installation, we need to go to the Flask Config API. Under config, we will get Mail-Server, Mail_Port, Mail_username, Mail_Password, etc options. We can use the mail.send() method to compose the message.

7. Do you have any idea about WSGI?

Ans: The full form of WSGI is Web Server Gateway Interface. This is actually a Python standard in PEP 3333. It provides the protocol for web servers to communicate with a web application. The current WSGI version is 1.0.1. It mainly plays a role at the time of project deployment.

8. Tell me the default local host and port in Flask?

Ans: The default Flask local host is 127.0.0.1 and the default port is 5000.

9. What do you know about Flask-wtf?

Ans: It provides an easy integration process with WTForms.

10. What features are available in Flask-wtf?

Ans: A few features are

1. Integration with WTForms.
2. CSRF token provides security globally.
3. Provide Recaptcha.
4. File upload facility.

11. How can we get a string in Flask?

Ans: We can get the string by using the argument’s value. This value will be used as the request object in Flask.

For example, we can try this sample code:

12. Why is Flask famous as a microframework?

Ans: It has some core features like requests, routing, and blueprints, as well as a few other features like coaching, ORM, and forms. Because of all these features, we call Flask a microframework.

13. How can we integrate any API like Facebook with the Flask application?

Ans: We can integrate it easily with the help of the Flask extension named Flask-Social. It gives multiple access to users for other social platforms also. One thing we need to remember is that we have to use the Flask-Security extension of Flask for security purposes. For this, we need to install all social API libraries in Python. We have to register from an external API service provider.

14. Can we use the SQLite database in Flask?

Ans: Yes, it is built-in with Python as a database. We don’t have to install any extensions. Inside the Python view, we can import SQLite. There we can write SQL queries to interact with a database. Generally, the Python Flask developers use Flask-SQLAlchemy, which makes SQL queries easier. It has one ORM, which helps to interact with the SQLite database.

15. Can you briefly talk about the Flask template engine?

Ans: The Flask template engine allows developers to create HTML templates with placeholders for dynamic data. Actually, a template is a file that contains two types of data, one is static and another is dynamic. Mostly this dynamic template is popular because the data is in run time. Flask allows the Jinja2 template engine to be used mostly as a template engine. Flask’s render_template method needs parameters and their values.

16. Explain the thread-local Flask object?

Ans: Thread local Flask objects are mostly available in a valid request context. Because of this, we don’t need to pass objects from one method to another. Because in Flask thread safety is declared out of the box. We can access the objects by a command like the current _app.

17. Tell me why we need to use Flask, not Django?

Ans: See, Flask is a better quick development Python framework. It uses cases and perfects prototyping. It is better for lightweight web applications. It is best to develop microservice and server-less applications. In comparison, it has full features that are built-in. Flask is also easy to learn with so many APIs. In Django, there are fewer APIs.

18. What are the features of forms extension in Flask?

Ans: We have to use the Flask- extension to implement forms in Flask. It’s called WTForms. It’s a Python-based rendering and validation library. It does data validation, CSRF protection, and internationalization. Once we use Flask-uploads, ReCaptcha from Flask-WTF helps to upload files. We can also manage JavaScript requests and customization of error responses.

19. Explain the G object in Flask?

Ans: If we want to hold any data during the application context, the Flask g object is used as a global namespace for keeping any data. It’s not suitable for storing data within requests. Actually, this letter g stands for global. Suppose we need to store a global variable in an application context, then g object is best in place of creating global variables. Here the g object will work as a request in a separate Flask g object. It always saves self-defined global variables.

20. Tell me about the application context in Flask?

Ans: It is the basic idea to complete the response circle in the Flask application. During any request of the CLI command, it maintains all track of application-level data. We can use g object or current _app to access those data. Always Flask forced application context with every request to complete the circle.

21. Tell me, how will you create the RESTFul application in the Flask framework?

Ans: There are so many extensions that we can use to create the RESTFul application in Flask. We need to choose them depending upon the requirements.

A few of them are

1. Flask-RESTFul.
2. Flask-API.
3. Flask-RESTX.
4. Connexion.

22. How can we get a user agent in Flask?

Ans: We can use the request object, see the given below code:

From flask import Flask
From flask import request

app=Flask(_name_)

@app.route(“/”)
Def index();
                val=request.args.get(“var”)
user_agent=request.headers.get(‘User-Agent’)

response=”””
&It;p>
Hello, World!{}
&It;/br/>
  you are accessing this app with {}
&It;p>
“””.format(val,user_agent)
return response
if_name_==”_main_”;
            app.run(host=”0.0.0.0”,port=8080)

23. Explain to me how to use URLs in Flask?

Ans: We need to call the view function with parameters and give them values to generate URLs. Mainly Flask url_for the function we use here. It can also be used in Flask templates.

24. Tell me, how will you create an admin interface in Flask?

Ans: Here we also need to use one of the Flask extensions named Flask-Admin. It helps to group all individual views together in classes. Another extension named Flask-Appbuilder can be used. It has a built-in admin interface.

25. Explain the process of using the session in Flask?

Ans: The session is mainly used to store data in requests. We can store and get data from the session object in Flask. As shown below code we can try to do this:

Fromflask import Flask,session

app=Flask(_name_)
@app.route(‘/use_session’)
Def use_session{}
                 If’song’not in session;
                 session[;songs’]={‘title’;’Tapestry’,’Bruno Major’}

Return session.get(‘songs’)

@app.route(‘/delete_session’)
def delete_session{}
                   session.pop(‘song’,None)
                   return “removed song from session”

26. Can we debug the Flask application?

Ans: Yes, we can debug the Flask application. Every development server has a debugging facility. Flask also comes with one server, so one server is there by default. If we run the method to call the Flask application object, we need to keep the debug mode value true. Try this given below code:

Remember that we need to deactivate the debug mode before deploying; otherwise, a full stack trace will be displayed in a browser. It is not secure as it has confidential details. One extension is also available named Flask-DebugToolbar.

27. Tell me the time of the identifier in Flask?

Ans: Actually, it can be any length. But there are certain rules which we must follow:

1. The identifier should start with a  character or an underscore, or from A-Z or a-z.
2. A-Z or a-z can be stored as a name in the identifier.
3. Python Flask is case-sensitive.
4. Few keywords are not usable in identifiers like and, false, import, true, del, try, etc.
5. How many HTTP methods can we use in Flask? Explain.

Basically, we use 5 types of HTTP methods to retrieve data from URLs. They are:

1. GET: It sends the unencrypted data to the server.
2. POST: Post server caches all data except HTML.
3. HEAD: It is similar to GET without any response body.
4. PUT: It can replace current data.
5. DELETE: It can delete the current data requested by any URL.

Related Article: Python vs Java – Differences

28. What is the procedure for database connection requests in Flask?

Ans: We can do it in three ways, they are:

1. after_request(): It helps make a request and passes the response, which will be sent to the client.
2. before-request(): It is called before the request without any argument passing.
3. teardown_request(): In case we get an exception, then this connection will be used and response is not guaranteed.

29. How can we create request context in the flask?

Ans: It can be created in two easy steps. They are:

1. It can be created on its own when the application receives a request from the system.
2. We can do it manually by calling app.test_request_context

30. Is Flask an MVC framework?

Ans: Yes, it is an MVC( Model View Control) framework. Because it has a feature named session, this helps to remember information from one request to another request. It uses a signed cookie to show the contents of that session to the user. If the user wants to modify. They have to use one secret key named Flask.secret_key in Flask. Flask perfectly behaves like one MVC framework.

31. Explain Flask Sijax?

Ans: It’s an inbuilt library in Python, making it easy for Ajax to work in web applications. Sijax uses JSON while passing data to a server called by the browser.

32. Tell me how to show all errors in the browser for Flask?

Ans: We need to run Python files on the shell. The command will be app. debug=True

33. Explain how we can structure a huge big flask application?

Ans: We need to follow the steps below to structure a huge big flask application:

1. We need to move the functions to different files until the applications get started.
2. We need to use the blueprint to view the categories like auth and profile.
3. We need to register all functions on a central URL map using the Werkzeug URL.

34. What is the utilization of jesonify() in a flask?

Ans: This is one of the functions under the flask.json module. It can convert data to JSON and store it in the response object. It provides a response object with an application where json.dumps() only returns a JSON data string.

35. Describe Flask memory management in short?

Ans: Memory management handles memory allocation in Python. Flask has a built-in garbage collector. It collects all wastage data and makes the space free. It manages memory by private heap space. This is not accessible by developers. Few of them are accessed by users by using a few API tools. All data allocation is done by Flask memory management.

36. What do you know about the validators class of WTForms in Flask?

Ans: A Validator can take the input to check if it meets some criteria like string limit with returns. In case of failure, a validation error will come. This is a straightforward method. In Flask there are a few validator classes of WTForms. They are:

1. DataRequired: Checks the input field.
2. Email: It checks the email id conventions.
3. IP Address: It checks the IP address.
4. Length: It validates the string length with the given range.
5. NumberRange: It validates the number in the input field with the given range.
6. URL: Checks the URL input field.

37. Can we get any visitor’s IP address in Flask? Explain how?

Ans: Yes, we can do it. Request.remo0te_addr is useful to get the IP address in Flask. An example is given below:

From flask import request
From flask import jsonify
@app.route("/get_user_ip",method=["GET"]
Def get_user_ip():
Return jsonify({'ip': request.remote_addr}), 200

38. Explain Flask error handlers?

Ans: Generally, an HTTP error code will be returned when one error comes. Suppose the error code is within 400 to 499; then it is sure that a mistake is happening in the client-side request. Otherwise, if the error code is within 500 to 599, the error comes from a server-side request.

HTTP error code can show custom error pages to the user. This HTTPS error code is not set to the error handler’s code. While returning a message from the error handler, we have to include it.

40. Write the code to change the default host and port in Flask?

Ans: We can try this code:

Fast API

 

Django

Tools