Unit 1: Supervised Learning Algorithms

Part A

1. What is machine learning? (Part A, Q1)

2. What is logistic regression? (Part A, Q2)

Part B

1. Give decision trees to represent the following Boolean functions: (Part B, Q1)

o (a) A ˄˜B

o (b) A V [B ˄ C]

o (c) A XOR B

o (d) [A ˄ B] v [C ˄ D]

2. What is Bayes theorem with an example? (Part B, Q2)

3. Define: (Part B, Q3)

o (i) Prior Probability

o (ii) Conditional Probability

o (iii) Posterior Probability

4. Describe K-nearest Neighbour learning algorithm for continuous-valued target function (Part
B, Q13)

5. Discuss the major drawbacks of K-nearest Neighbour learning algorithm and how it can be
corrected (Part B, Q14)

6. Define the following terms with respect to K-Nearest Neighbour Learning: (Part B, Q15)

o Regression

o Residual

o Kernel Function

7. Explain the K-nearest neighbour algorithm for approximating a discrete-valued target

8. What is logistic regression and linear regression? Explain the difference between them (Part
B, Q19)

9. What is Support Vector Machine Algorithm? Explain the working of SVM (Part B, Q25)

Part C

1. Why is the Naive Bayesian classification called “naive”? Explain (Part C, Q1)

2. The support vector machine (SVM) is a highly accurate classification method. However, SVM
classifiers suffer from slow processing when training with a large set of data tuples. Describe
 how to overcome this difficulty and design a scalable SVM algorithm for efficient SVM
classification in large datasets (Part C, Q2)

Unit 2: Unsupervised Learning Algorithms

Part A

1. What are types of clustering methods? (Part A, Q3)

2. Explain Clustering and Association (Part A, Q11)

3. What is hierarchical Clustering in Machine Learning? (Part A, Q12)

Part B

1. What is K-Mean Algorithm? Explain with help of example (Part B, Q20)

2. How to measure the distance between elements of the cluster? Explain different methods to
measure the distance (Part B, Q21)

3. Define Hierarchical Clustering. Explain Agglomerative and Divisive hierarchical clustering

Part C

1. Give an example to show that items in a strong association rule actually may be negatively
correlated (Part C, Q3)

2. Explain the k-Means Algorithm with an example (Part C, Q4)

3. Define Hierarchical Clustering. Explain Agglomerative and Divisive hierarchical clustering

Unit 3: Statistical Learning Theory, Feature Extraction, and Feature Selection

Part A

1. Define singular value decompositions (Part A, Q6)

2. Define Feature Extraction and Feature Selection (Part A, Q13)

3. What is Singular Value Decomposition? (Part A, Q14)

4. What is Hashing? Explain (Part A, Q16)

Part B

1. Differentiate between feature extraction and feature selection (Part B, Q5)

2. What is Singular Value Decomposition with example? (Part B, Q6)

3. What is Principal Component Analysis? Explain step-by-step computations of PCA (Part B,

4. Explain the following feature subset selection methods: (Part B, Q27)

o Wrapper Method

o Embedded Method

Part C

1. Explain the concept of Principal Component Analysis with a suitable example (Part C, Q5)

2. Explain the following feature subset selection methods: (Part C, Q13)

o Filter Method

o Wrapper Method

o Embedded Method

Unit 4: Semi-Supervised and Reinforcement Learning

Part A

1. Express the Markov property mathematically (Part A, Q7)

2. Give clear difference between episodic and continuous tasks of Markov process (Part A, Q8)

3. What is Markov decision process? Explain (Part A, Q15)

Part B

1. Explain the Q function and Q Learning Algorithm (Part B, Q7)

2. Explain the concept of Markov Decision Process (Part B, Q8)

3. What is Reinforcement Learning? (Part B, Q11)

4. Explain Q learning algorithm assuming deterministic rewards and actions (Part B, Q16)

5. What is Bellman Equation? Explain (Part B, Q23)

Part C

1. What is the Markov decision process? Explain it with suitable example (Part C, Q7)

2. Explain various methods of SARSA (Part C, Q8)

3. What is Q-Learning? Explain with help of example (Part C, Q14)

Unit 5: Recommendation Systems, Neural Networks, and Deep Learning

Part A

1. What is deep learning? (Part A, Q9)

2. What is Artificial Neural Network? (Part A, Q10)

3. What is Recommended System? Define it (Part A, Q17)

Part B

1. What is Artificial Neural Network? What are the types of problems in which Artificial Neural
Networks can be applied? (Part B, Q9)

2. Explain the concept of a Perceptron with a neat diagram (Part B, Q10)

3. Explain the Q function and Q Learning Algorithm (duplicate question in Unit 4) (Part B, Q12)

4. What is Perceptron? Explain the basic components of perceptron (Part B, Q24)

5. Define Backpropagation. Explain working of backpropagation algorithm (Part B, Q30)

Part C

1. Explain the role of collaborative-based and content-based recommendation systems in

2. How does the collaborative recommendation system differ from a typical classification or
predictive modeling system? Explain (Part C, Q10)

3. Define Backpropagation. Explain working of backpropagation algorithm (Part C, Q15)