QUESTION BANK MACHINE LEARNING (ROE083)

Department of Electronics and Communication Engineering

COURSE B.Tech., ECE, 8th SEM,

Unit-1 Introduction to ML & Concept Learning

Unit -1 Syllabus: INTRODUCTION – Well defined learning problems, designing a Learning

System, Issues in Machine Learning; THE CONCEPT LEARNING TASK - General-to-specific
ordering of hypotheses, Find-S, List then eliminate algorithm, Candidate elimination algorithm,
Inductive bias.

1. What is Machine Learning:

a) The autonomous acquisition of knowledge through the use of computer programs

b) The autonomous acquisition of knowledge through the use of manual programs

c) The selective acquisition of knowledge through the use of computer programs

d) The selective acquisition of knowledge through the use of manual programs

Ans: a) The autonomous acquisition of knowledge through the use of computer programs

2. Different learning methods does not include:

(a) Memorization

(b) Analogy

(c) Deduction

(d) Introduction

Ans: (d) Introduction

3. Training is required in machine learning for

(a) to build a successful model

(b) to build an algorithm

(c) to make machine learn

(d) all

Ans: (d) all

4. The performance measure of a Robot Car driving Learning problem is______________

(a) Distance travelled before an error

(b) distance travelled after an error

(c) No of turns made by car

(d) Zigzag path followed distance

Ans: (a) Distance travelled before an error

5. What are the features of a well-defined learning problem?

(a) Task, Performance, Training Experience

(b) Classification & Regression

(c) Probability, Events, Inputs-Outputs

(d) None

Ans: (a) Task, Performance, Training Experience

6. Artificial Intelligence is small as compared to Machine Learning (Yes/No)

Ans: No

11. Explain different approaches in Machine Learning are:

(a) Concept Versus Supervised Learning

(b) Symbolic vs Statistical learning

(c) Inductive Vs Unsupervised Learning

(d) All

Ans: (d) All

7. Contrast what is not Machine Learning?

(a) Artificial Intelligence

(b) Rule-based inference

(c) Both

Ans: (a) Artificial Intelligence

8. Explain what is the function of ‘Unsupervised Learning’?

(a) Find clusters of the data

(b) Find low-dimensional representations of the data

(c) Find interesting directions in data

(d) All the above

Ans: (a) Find clusters of the data

9. Explain what is the function of ‘Supervised Learning’?

(a) Classifications

(b) Clustering

(c) Sorting

Ans: (a) Classifications

10. Explain what is algorithm independent machine learning?

a) where mathematical foundations is independent of any particular classifier

b) where mathematical foundations is dependent of any particular classifier

Ans: a) where mathematical foundations is independent of any particular classifier

11. Explain what is classifier in machine learning?

A classifier in a Machine Learning is a system that inputs a vector of discrete or continuous feature
values and outputs a single discrete value, the class.

a) The above statement is false

b) The above statement is true

Ans: b) The above statement is true

12. Explain what is Model Selection in Machine Learning?

a) selecting models among different mathematical models, which are used to describe the
same data set

b) selecting models among different mathematical models, which are used to describe the
different data set

c) All the above

Ans: a) selecting models among different mathematical models, which are used to describe the
same data set

13. what do you understand by eigenvectors:

(a) Whose direction whose direction remains unchanged when a linear transformation is applied to
it
(b) Whose direction whose direction is changed when a linear transformation is applied to it

Ans: (a) Whose direction whose direction remains unchanged when a linear transformation is
applied to it.

14. Which of the following are comprised within AI:

(a) Machine Learning

b) Deep Learning

c) Both (1) and (2)

d) None of the above

Ans: c) Both (1) and (2)

15. Which of the mentioned human behavior does the AI aim to mimic?

a) Thinking

b) Eating

c) Sleeping

d) None of the above

Ans: a) Thinking

16. Which of the following is not a goal of AI?

a) Thinking humanly

b) Adapting to the environment and situations

c) To rule over humans

d) Real Life Problem Solving

Ans: c) To rule over humans

17. Identify that to find the minimum or the maximum of a function, we set the gradient to zero
because:
a. The value of the gradient at extrema of a function is always zero

b. Depends on the type of problem

c. Both A and B

d. None of the above

Ans: a. The value of the gradient at extrema of a function is always zero

18. Explain Find S Algorithm is an Example of:

a) Concept learning

b) Reinforcement learning

c) Supervised learning

d) Unsupervised Learning

Ans: a) Concept learning

19. Explain the different Algorithm techniques in Machine Learning?

a) Supervised Learning

b) Unsupervised Learning

c) Reinforcement Learning

d) All the above

Ans. d) All the above

20. A computer program is said to learn from experience E with respect to some class of tasks T and
performance measure P, if its performance at tasks in T, as measured by P, improves with
experience E. (True/False)

Ans: True

21. The Machine Learning is the autonomous acquisition of knowledge through the use of computer
programs.(True/False)

Ans: True
22. A checkers game is not a well-defined learning problem. (True/False)

Ans: False

23. Which attribute is best classifier : (a) Humidity , (b) Wind, (c) Both (d) None

Ans: Humidity

24. A classifier in a Machine Learning is a system that inputs a vector of discrete or continuous
feature values and outputs a _____________

a) Continues Value

b) Discrete value

c) Both

d) None

Ans: b) Discrete value

25.

a) Supervised Classification

b) Unsupervised classification

c) Regression
d) Model

Ans: a) Supervised Classification

26.A feature F1 can take certain value: A, B, C, D, E, & F and represents grade
of students from a college.
Which of the following statement is true in following case?
A) Feature F1 is an example of nominal variable.
B) Feature F1 is an example of ordinal variable.
C) It doesn’t belong to any of the above category.
D) Both of these
Answer- (B)
 Unit-2 Decision Tree & ANN

Unit -2 Syllabus: DECISION TREE LEARNING - Decision tree learning algorithm-Inductive bias-
Issues in Decision tree learning; ARTIFICIAL NEURAL NETWORKS – Perceptrons, Gradient
descent and the Delta rule, Adaline, Multilayer networks, Derivation of backpropagation rule
Backpropagation Algorithm Convergence, Generalization;

27. Which of the following is a disadvantage of decision trees?

(a) Factor analysis

(b) Decision trees are robust to outliers

(c) Decision trees are prone to be overfit

(d) None of the above

Ans: (c) Decision trees are prone to be overfit.

28. The hidden layer in Artificial neural network(ANN) is _______________

(a) Input Layer

(b) Middle Layer

(c) Output Layer

(d) All Layer

Ans: (b) Middle Layer

29. Sigmoid function is a type of ________________

(a) Linear Function

(b) Activation Function

(c) Input Function

(d) output Function

Ans: (b) Activation Function

30. Modify the statement and find the missing word. In Machine Learning, Perceptron is an
algorithm for ……………….of the input into one of several possible non-binary outputs.
Fill in the blank

a) Unsupervised learning

b) supervised classification

c) Artificial Neural Network

d) All the above

Ans: c) Artificial Neural Network

31. Select from the given option .What is the use of Sigmoid Function

a) To Predict the probability as an output

b) To Predict the probability as an input

c) To Predict the probability as an input and output

d) All three options are correct

ANs: a) To Predict the probability as an output.

32. In Machine Learning SVM Stands for ____________

(a) Service Vector Machine

(b) Support Vector Machine

(c) Sector Vision Memory

(d) All options

Ans: (b) Support Vector Machine

33. Which of the following is an example of a deterministic algorithm?

A) PCA

B) K-Means

C) None of the above

D) All

Ans: A) PCA
34. In Machine Learning, Iterative Dichotomize (ID3) algorithm is used to generate a decision tree
from a dataset.(True/False)

Ans: True

35. Which of the following hyper parameter(s), when increased may cause
random forest to over fit the data?
1. Number of Trees
2. Depth of Tree
3. Learning Rate
A) Only 1
B) Only 2
C) Only 3
D) 1 and 2
E) 2 and 3
F) 1,2 and 3
Answer-(b)

36. Imagine, you are working with “Analytics Vidhya” and you want to develop
a machine learning algorithm which predicts the number of views on the
articles. Your analysis is based on features like author name, number of articles
written by the same author on Analytics Vidhya in past and a few other features.
Which of the following evaluation metric would you choose in that case?
1. Mean Square Error
2. Accuracy
3. F1 Score
A) Only 1
B) Only 2
C) Only 3
D) 1 and 3
E) 2 and 3
F) 1 and 2
Answer-(A)

37. Let’s say, you are using activation function X in hidden layers of neural
network. At a particular neuron for any given input, you get the output as “-
0.0001”. Which of the following activation function could X represent?
A) ReLU
B) tanh
C) SIGMOID
D) None of these
Answer-(B)

38. Choose the appropriate word, When there is sufficient data ‘Isotonic
Regression’ is used to prevent an……………...
Fill in the Blank
a) overfitting issue
b) Under fitting issues
c) both

Answer-(A)

39. A perceptron in ANN is equivalent to Neuron in Human Brain.

(True/False)
Answer-( True)
 Unit-3 Hypothesis & Bayesian Learning

Unit -3 Syllabus: Evaluating Hypotheses: Estimating Hypotheses Accuracy, Basics of sampling

Theory, Comparing Learning Algorithms; Bayesian Learning: Bayes theorem, Concept learning,
Bayes Optimal Classifier, Naïve Bayes classifier, Bayesian belief networks, EM algorithm;

40. The most general hypothesis-that every day is a positive example-is represented

By :

(a)(??????),

(b) (000000),

(c) (111111),

(d) (ØØØØØØ)

Ans: (a)(??????)

41. A Bayesian Belief Network is a Type of _________________________

(a) Probabilistic Graph Model

(b) Decision Tree

(c) Network Model

(d) None

Ans. (a) Probabilistic Graph Model

42. The process of evaluating a given hypothesis is called _________________

(a) Hypothesis formation

(b) Hypothesis Testing

(c) Both

(d) None

Ans: (b) Hypothesis Testing

43. Explain what are the advantages of Naive Bayes?

Option 1: In Naïve Bayes classifier will converge quicker than discriminative models like logistic
regression, so you need less training data. The main advantage is that it can’t learn interactions
between features.

Option 2: In Naïve Bayes classifier will converge quicker than discriminative models like logistic
regression, so you need More training data. The main advantage is that it can’t learn interactions
between features.

a) Both options are true

b) Option 1 is true

c) Option 2 is true

Ans: b) Option 1 is true

44. List the most widely used metrics and tools to assess a classification model are:

a. Confusion matrix

b. Cost-sensitive accuracy

c. Area under the ROC curve

d. All of the above

Ans: d. All of the above

45. Show the one option which best suits .PAC …………………….learning is a learning framework that
has been introduced to analyze learning algorithms and their statistical efficiency.

Fill in the blank

a) Probably Approximately Correct

b) Probably Approximate Correct

c) Probably Approximately Correctly

d) All the above

Ans: a) Probably Approximately Correct

46. In Machine Learning, a Hypothesis is an assumption about a fact of an entity. (True/False)

Ans: True

47. In Machine Learning, the Confusion Matrix is not used for performance test of an algorithm.
(True/False)

Ans: False

48. Given a hypothesis space H, a hypothesis (h Є H) is said to over fit the Training data, if there
exists some alternative hypothesis (h' Є H), such that h has smaller error than h' over the training
examples, but h' has a smaller error than h over the entire distribution of instances.(True/False)

Ans: True

49. The BACKPROPAGATION algorithm is not used in ANN. (True/False)

Ans: False

50. The true error of a hypothesis is the probability that it will misclassify a

single randomly drawn instance from the distribution D.(True/False)

Ans: True

51. Bayes theorem is the cornerstone of Bayesian learning methods because

it provides a way to calculate the posterior probability P(hlD), from the prior

Probability P(h), together with P(D) and P(D(h).(True/False)

Ans: True

52. Concept learning is Inferring a Boolean-valued function from training examples of its input and
output. (True/False)

Ans: True

53. The VERSION SPACE (VSHD) with respect to hypothesis space H

and training examples D, is the subset of hypotheses from H consistent with the

training examples in D.(True/False)

Ans: True

54. Bayes theorem is the cornerstone of Bayesian learning methods because

it provides a way to calculate the posterior probability P(hlD), from the prior

Probability P(h), together with P(D) and P(D(h).(True/False)

Ans: True

55. VC dimension is a measure of _________________ of a Hypothesis space (H) which can be learnt
by a classification algorithm.

(a) Error

(b) Complexity

(c) Memory

(d) iterations

Ans: (b) Complexity

56. Describe What are the three stages to build the hypotheses or model in
machine learning?
(a) Model building, Model testing
(b) Applying the model
(c) All the above

Answer-(A)

57. Choose the following missing word from Bayesian Network where it is
used to represent the ……………….for probability relationship among a set of
variables.
Fill in the blank
a) Mathematical Model
b) Stochastic Model
c) graphical model
d) All the above

Answer-(C)
58. Option1: Hypothesis space is the set of all the possible legal hypothesis.
This is the set from which the machine learning algorithm would
determine the best possible (only one) which would best describe the
target function or the outputs.
Option 2: A hypothesis is a function that best describes the target in
supervised machine learning. The hypothesis that an algorithm would
come up depends upon the data and also depends upon the restrictions
and bias that we have imposed on the data.
Both are correct
Option 2 is correct
Both are incorrect
Option one is correct
Answer-(D)

59. Bayes theorem is the cornerstone of Bayesian learning methods because

it provides a way to calculate the posterior probability P(hlD), from the
prior Probability P(h), together with P(D) and P(D(h).
(True/False)

Answer-(True)

60. What is a top-down parser?

a) Begins by hypothesizing a sentence (the symbol S) and successively predicting lower level
constituents until individual preterminal symbols are written

b) Begins by hypothesizing a sentence (the symbol S) and successively predicting upper level
constituents until individual preterminal symbols are written

c) Begins by hypothesizing lower level constituents and successively predicting a sentence (the
symbol S)

d) Begins by hypothesizing upper level constituents and successively predicting a sentence (the
symbol S)

Answer: a
 Unit-4 CLT & k-NN

Unit -4 Syllabus: Computational Learning Theory: Sample Complexity for Finite Hypothesis
spaces, Sample Complexity for Infinite Hypothesis spaces, The Mistake Bound Model of
Learning; INSTANCE-BASED LEARNING – k-Nearest Neighbour Learning, Locally Weighted
Regression, Radial basis function networks, Casebased learning

61. The CLT stands for_____________________

a) Computational Learning Technology

b) Computer Learning Technique

c) Computer Learning Technology

d) Computational Learning Technique

Ans. a) Computational Learning Technique

62. Contrast In what areas Pattern Recognition is used?

Pattern Recognition can be used in

(a) Computer Vision

(b) Speech Recognition

(c) Data Mining

(d) All the above

Ans: (d) All the above

63. Show which of the following is a widely used and effective machine learning algorithm based on
the idea of bagging?

a. Decision Tree

b. Regression

c. Classification

d. Random Forest

Ans: d. Random Forest

64. Show the one option which best suits .Euclidian distance is the part of :

a) Supervised Learning

b) KNN Algorithm

c) Genetic Algorithm

d) All the above

Ans: b) KNN Algorithm

65. Show the one option which best suits .What is the full form of CLT

a) Computational Learning Technology

b) Computer Learning Technique

c) Computer Learning Technology

d) Computational Learning Technique

Ans: d) Computational Learning Technique

 Unit-5 Genetic Algorithm & Reinforcement Learning

Unit -5 Syllabus: Genetic Algorithms: an illustrative example, Hypothesis space search, Genetic
Programming, Models of Evolution and Learning; Learning first order rules sequential covering
algorithms-General to specific beam search-FOIL; REINFORCEMENT LEARNING - The Learning
Task, Q Learning

66. Explain what is Genetic Programming?

Option 1: Genetic programming is one of the two techniques used in machine learning. The model is
based on the testing and selecting the best choice among a set of results.

Option 2: Genetic programming is one of the two techniques used in machine learning. The model is
based on the testing and selecting the best choice among a set of results and is used for optimization

a) Only Option 1 is correct

b) Only option 2 is correct

c) Both are Correct

d) Both are Incorrect

Ans: c) Both are Correct

67. Explain what is Inductive Logic Programming in Machine Learning?

a) Logical programming representing background knowledge and examples.

b) Logical Algorithm representing background knowledge and examples.

Ans: a) Logical programming representing background knowledge and examples.

68. The action ‘STACK(A, B)’ of a robot arm specify to _______________

a) Place block B on Block A

b) Place blocks A, B on the table in that order

c) Place blocks B, A on the table in that order

d) Place block A on block B

Ans: d) Place block A on block B.

69. An AI agent is defined through its PEAS. What does the term PEAS stands for:

(a) Personal Enhancement Area in Science

(b) Performance, Environment, Actuators and Sensors
(c) Performance, Entity, Area, State
(d) None of the above

Ans: (b) Performance, Environment, Actuators and Sensors

70. "In AI, we study the whole universe by dividing it into two components." What are these two
components?

a) Sky and Land

b) Agent and environment

c) Yes or No

d) None of the above

Ans: b) Agent and environment

71. Show the one option which best suits .Cross over is the part of?

a) KNN Algorithm

b) Radial Basis Function

c) Genetic Algorithm

d) All the above

Ans: c) Genetic Algorithm

72. Select from the given option .Reinforcement Learning is the ………………… Based Learning

Fill in the blank

a) Award based

b) Model based

c) Reward based

d) All the above

Ans: c) Reward based

73. Select from the given option . Q- learning is the part of

a) Supervised learning

b) Unsupervised learning

c) Natural Learning

d) Reinforcement Learning

Ans: d) Reinforcement Learning

74. Select from the given option. Explain Reward Maximization

a) Takes the best action so that the reward is maximum

b) Takes the moderate action to maximize the reward

c) All the above

Ans: a) Takes the best action so that the reward is maximum

75. Select from the given option . What is the Full form of MDP

a) Marks Decision procedure

b) Markov’s Decision Process

c) A) and b) is false

d) A) and b) is true

Ans: b) Markov’s Decision Process

76. Select from the given option .Mutation is the process involved in

a) Find S algorithm

b) K- means Clustering

c) Genetic Algorithm

d) All the above

Ans: c) Genetic Algorithm

77. Show the one option which best suits .Q" names the function that returns the reward used to
provide the reinforcement and can be said to stand for the …………of an action taken in a given state.
Fill in the blank

a) "quality"

b) Quantity

c) Both

Ans: a) "quality"

78. Inductive learning algorithms are not able to classify unseen examples only because of their
implicit inductive bias for selecting one consistent hypothesis Over another. (True/False)

Ans: False

79. An Agent is used in ______________

(a) Detective

(b) Machine Learning

(c) Reinforcement Learning

(d) ANN

Ans: (c) Reinforcement Learning

80. Which of the mentioned parts of an agent would you consider to be the most valuable in terms of
AI?
a) Sensors and Actuators
b) Wheels and steering
c) Arms and legs
d) All of the above
Ans. a) Sensors and Actuators