Questions Bank

Regulation 2019
III-Year-VI Semester

1903BS005: MACHINE LEARNING

MODULE-1
Q. No [CO#, BTL]
PART A – (2 marks)
1. How do you define machine learning? Enumerate its importances. [CO1, L2]
2. State the importance of hypothesis in Machine Learning. [CO1, L2]
3. List the applications of machine learning with examples. [CO1, L2]
4. Why do you need the feature engineering? [CO1, L2]
5. Elaborate the use of PAC learning model. [CO1, L2]
Q. No PART B – (16 marks) [CO#, BTL]
1 Explain in brief about the feature engineering techniques in machine learning with suitable examples. [CO1, L2]
2 Discuss the PAC learning model with illustrations. [CO1, L2]
3 Illustrate the concepts of VC (Vapnik-Chervonenkis) dimension for linear and rectangular classifier [CO1, L2]
with examples and neat diagram.
MODULE-2
PART A – (2 marks)
1. Where do you apply the conditional probability? [CO2, L2]
2. Compare discrete random variable and continuous random variable. [CO2, L2]
3. How do you understand the reinforcement Learning? [CO2, L2]
4. What do you mean the imbalanced data? [CO2, L2]
5. Why should we detect outliers? [CO2, L2]
Q. No PART B – (16 marks) [CO#, BTL]
1 Why should we detect outliers? How will you detect outliers using standard deviation? [CO2, L2]
2 Describe the following machine learning Algorithms with neat diagram and examples in detail. [CO2, L2]
(i)Supervised Learning
(ii)Unsupervised Learning.
(iii)Reinforcement Learning
3 Elaborate the concepts of Probability distributions in machine learning algorithms with neat diagram [CO2, L2]
and examples.

MODULE-3
Q. No [CO#, BTL]
PART A – (2 marks)
1. Outline the need of Bayesian Belief Network (BBN). [CO3, L2]
2. State cross validation of dataset of a model. [CO3, L2]
3. List the various model evaluation methods. [CO3, L2]
4. How do you state Area Under Curve (AUC) of model? [CO3, L2]
5. Determine the training and testing classification model. [CO3, L2]
Q. No PART B – (16 marks) [CO#, BTL]
1 Explain about Naive Bayes classification algorithm for continuous attributes with examples. [CO3, L2]
 2 Illustrate the concepts of K-Nearest Neighbor (KNN) classification algorithm with neat diagram. [CO3, L2]
3 Explain the need of Support Vector Machines (SVM) with neat diagram in detail. [CO3, L2]

MODULE-4
Q. No [CO#, BTL]
PART A – (2 marks)
1. Give the applications of classifications. [CO4, L2]
2. Compare bagging and boosting. [CO4, L2]
3. Why do you need Artificial Neural Networks (ANN)? [CO4, L2]
4. Illustrate the use of Hidden Markov Models (HMM). [CO4, L2]
5. Differentiate ANN with HNN. [CO4, L2]
Q. No PART B – (16 marks) [CO#, BTL]
1 Illustrate the ensembles of classifiers including bagging and boosting with example in detail. [CO4, L2]
2 Describe the Hidden Markov Model (HMM) with following algorithms. [CO4, L2]
(i) forward-backward algorithm.
(ii)Viterbi algorithm
3 Enumerate the applications of sequence classification such as part-of -speech (POS) tagging with neat [CO4, L2]
diagram.
MODULE-5
Q. No [CO#, BTL]
PART A – (2 marks)
1. When to use the regression? [CO5, L2]
2. State outlier detection. [CO5, L2]
3. List the applications of regression. [CO5, L2]
4. Mention the need of least squares regression. [CO5, L2]
5. Give the importance of LASSO. [CO5, L2]
Q. No PART B – (16 marks) [CO#, BTL]
1 Compare linear vs nonlinear regression and explain back propagation nonlinear regression in detail. [CO5, L2]
2 Illustrate the need of association rule mining algorithms including apriori with neat diagram. [CO5, L2]
3 Enumerate the concepts of Expectation-Maximization (EM) algorithm for unsupervised learning with [CO5, L2]
neat diagram.

CO – Course Outcomes, BTL - Bloom’s Taxonomy Levels (L1 – Remember, L2 –