End Semester Examination, April – May, 2025
Degree: [Link], Semester: Fourth
Stream: CSE(AI & ML)/ CSE(IOT, CS, BT)/
CSE/CSIT/CST/CSE(AI)/CSE(IoT)/ CSBS/IT
Paper Code: PCCCS403
Paper Name: Artificial Intelligence & Machine Learning
Full Marks: 100 Duration: 3 Hours
Part - A
Attempt any 10 questions out of 15 questions
Each question carries 2 Marks (2 × 10)
1. What is the Turing Test? 2
2. What is a knowledge base in AI? 2
3. What is game theory in AI? 2
4. Differentiate between single-agent and multi-agent task environment. 2
5. What is the difference between a state space and a search tree? 2
6. Define goal-based agent. 2
7. What is overfitting? How can it be avoided? 2
8. What is difference between Normalization and Standardization? 2
9. Differentiate between Bagging and Boosting in Ensemble Learning. 2
10. What is the role of regression model in exploratory data analysis? 2
11. How does the structure of a decision tree help in classifying a data instance? 2
12. Write the procedure for agglomerative clustering algorithm. 2
13. What is a subspace? Why is it important? 2
14. Define the terms agent and environment in the context of reinforcement 2
Page 1 of 4
learning.
15. What is Q-learning and what does the “Q” represent? 2
Part - B
Attempt any 6 questions out of 9 questions
Each question carries 5 Marks (5 × 6)
16. Compare depth-first search and breadth-first search. 5
17. Mention some of the challenges and limitations of search algorithms. 5
18. Can BFS be used for solving puzzles like the 8-puzzle problem? Why or why 5
not?
19. Explain the Minimax algorithm in game theory. 5
20. Explain how an imbalanced dataset can impact model performance. Suggest two 5
strategies to handle it.
21. Relate Entropy and Information Gain. 5
22. Explain the process of obtaining principle components and its relevance in 5
feature reduction.
23. Find cosine similarity, SMC and Jaccard coefficients for the following binary 5
data:
a) (1011) and (1100)
b) (10001) and (100001) and (11000)
24. What are some of the main challenges and issues faced in applying 5
reinforcement learning to real-world problems?
Part - C
Attempt any 5 questions out of 8 questions
Each question carries 10 Marks (10 × 5)
25. Discuss the importance of logical reasoning in AI and compare forward and 10
backward chaining.
26. Write pseudocode agent programs for the goal-based and utility-based agents. 10
27. Explain how the A* algorithm balances completeness and optimality while 10
searching for the shortest path in a graph.
Page 2 of 4
28. What is the entropy of this collection of training examples with respect to the 10
target function classification?
Instance Classification a1 a2
1 + T T
2 + T T
3 - T F
4 + F F
5 - F T
6 - F T
29. Consider an example of predicting a student’s result using Gaussian Naïve 10
Bayes algorithm for continuous attribute. The training dataset T consists of 10
data instances with attributes such as ‘Assessment Marks’, ‘Assignment Marks’
and ‘Seminar Done’ as shown in Table. The target variable is Result which is
classified as Pass or Fail for a candidate student.
Given a test data to be (Assessment Marks = 75, Assignment Marks = 6,
Seminar Done= Poor), predict the result of the student.
Table: Training Dataset
[Link]. Assessment Marks Assignment Marks Seminar Done Result
1. 95 8 Good Pass
2. 71 5 Poor Fail
3. 93 9 Good Pass
4. 62 4 Poor Fail
5. 81 9 Good Pass
6. 93 8.5 Poor Pass
7. 65 9 Good Pass
8. 45 3 Poor Fail
9. 78 8.5 Good Pass
10. 56 4 Poor Fail
30. Consider the following dataset describing customers of an electronic goods 10
shop. Classify the data X = (age = youth, income = medium, student = yes,
credit_rating = fair) using Naïve Bayes Classifier. Show calculations for priors
and likelihoods?
ID age student Class: Buys Computer
1 youth no no
2 youth no no
3 middle aged no yes
4 senior no yes
5 senior yes yes
6 senior yes no
Page 3 of 4
7 middle aged yes yes
8 youth no no
9 youth yes yes
10 senior yes yes
11 youth yes yes
12 middle aged no yes
13 middle aged yes yes
14 senior no no
31. Use the given Table dataset and apply hierarchical methods with seed points 10
(7,8) and (16,9). Show the dendrogram.
Table: Sample Data
[Link]. X Y
1. 3 5
2. 7 8
3. 12 5
4. 16 9
5. 20 8
32. Discuss the exploration-exploitation dilemma in reinforcement learning. Explain 10
why it is a challenging problem and describe several strategies that can be used
to balance exploration and exploitation. Provide examples of how these
strategies might be implemented.
**********
Page 4 of 4