IT-504-A Artificial Intelligence - Dec 2024 Question Paper

Under each question, How to answer lists the points an answer should cover, outlined by CampusPrep from the syllabus. Check the details against your textbook.

1. 7 Marks
   aExplain breadth first search algorithm. Give an example of a problem for which breadth first search would work better than depth first search.
   How to answer

   • Definition of BFS using queue, level-by-level expansion
   • Algorithm steps / pseudocode and time-space complexity
   • Example problem e.g. shortest path in unweighted graph / shallow goal
   • Comparison why BFS better than DFS for shallowest / optimal solution
   • Diagram of example search tree
2. 7 Marks
   bFind the shortest path using A* algorithm. Where

   

   How to answer

   • Note heuristic H(n) and edge costs g(n)
   • Compute f(n)=g(n)+h(n) iteration-wise with open/closed lists
   • Expand minimum f(n) node from S to G
   • Trace optimal path and total cost
   • Conclude shortest path found by A*
3. 7 Marks
   aApply resolution theorem to prove the following theorem. $P \lor Q, P \to R, Q \to R \Rightarrow R$
   How to answer

   • Convert premises to clause form and negate conclusion ~R
   • Show clauses: (P v Q), (~P v R), (~Q v R)
   • Apply resolution steps to derive contradiction / empty clause
   • Conclude R follows
   • State resolution theorem principle
4. 7 Marks
   bRewrite the following sentences in first order logic

   1. i)Coconut crunchy is a Biscuit.
   2. ii)Mary is a child who takes Coconut crunchy.
   3. iii)John loves children who take biscuits.
   4. iv)Jay loves Kites.
   How to answer

   • Identify constants, predicates: Biscuit, Child, Takes, Loves
   • Use quantifiers and connectives for each sentence
   • Distinguish specific individuals Mary, John, Jay, Coconut crunchy
   • Write well-formed FOL formulas i)-iv)
   • Maintain consistency of predicate names
5. 7 Marks
   aRepresent each of the following sentences of knowledge by a semantic net

   1. i)loves (mary, john)
   2. ii)loves (mary, john) $\land$ Hates (john, mita)
   3. iii)loves (mary, john) $\to$ Hates (mita, john)
   How to answer

   • Define nodes for Mary, John, Mita and relation links
   • Draw semantic net for loves(mary,john)
   • Add Hates link for conjunction case
   • Show implication link for conditional case
   • Label arcs with relation names
6. 7 Marks
   bConsider the following set of statements. Make a Bayesian network by constructing necessary conditional probability matrix.

   1. i)Patient has spots.
   2. ii)Patient has measles.
   3. iii)Patient has high fever.
   4. iv)Patient has rocky mountain spotted fever.
   5. v)Patient has previously been inoculated against measles.
   6. vi)Patient was recently bitten by a tick.
   7. vii)Patient has an allergy.
   How to answer

   • Identify causal variables: measles, inoculation, tick bite, allergy, fevers, spots
   • Draw DAG with dependencies and conditional independence
   • Construct conditional probability tables for each node
   • Explain diagnostic inference from symptoms
   • Show example CPT structure
7. 7 Marks
   aConsider the following blocks world problem. Start : $ON(C,B) \land ON(D,A) \land ONTABLE(B) \land ONTABLE(A) \land ARMEMPTY$ GOAL : $ON(C,B) \land ON(D,A) \land ONTABLE(B) \land ONTABLE(A)$

   1. i)Show how STRIPS would solve this problem.
   2. ii)Show how TWEAK would solve this problem.

   

   How to answer

   • Describe Start state ON(C,B), ON(D,A), ONTABLE(B), ONTABLE(A), ARMEMPTY and GOAL
   • Show STRIPS operators: Pickup, Putdown, Stack, Unstack with preconditions/add/delete
   • Give linear plan solving the blocks problem
   • Show TWEAK partial-order planning with constraints and threats
   • Compare linear vs non-linear planning
8. 7 Marks
   bHow would the minimax procedure have to be modified to be used by a program playing a three or four person game rather than a two person one?
   How to answer

   • Recall two-player minimax with max-min levels
   • Need vector evaluation for 3-4 players, each maximizing own utility
   • Propagate utility tuples up the game tree
   • Explain alliances / no single opponent assumption
   • Discuss pruning limitations
9. 7 Marks
   aGive one domain in which the expert system approach would be more promising and neural network approach would also be more promising.
   How to answer

   • Define expert system vs neural network strengths
   • Give domain e.g. medical diagnosis / rule-based vs pattern recognition
   • Justify why expert system promising: explicit rules, explanation
   • Justify why neural network promising: learning from data, noisy inputs
   • Conclude with example
10. 7 Marks
    bExplain basic model of ANN. What is the difference between biological neurons and artificial neural net?
    How to answer

    • Describe basic ANN model: inputs, weights, summation, activation, output
    • Give neuron output formula and activation functions
    • Draw artificial neuron vs biological neuron diagram
    • Compare: dendrites/axon/synapse vs inputs/outputs/weights
    • List differences in learning, speed, fault tolerance
11. 7 Marks
    aDefine Hill climbing algorithm with its limitations and their solutions.
    How to answer

    • Define hill climbing as local search, greedy ascent
    • Explain algorithm steps with current state evaluation
    • List limitations: local maxima, plateau, ridge
    • Give solutions: backtracking, random restart, simulated annealing
    • Illustrate with state-space diagram
12. 7 Marks
    bWhat is knowledge representation? Explain its approaches and issues.
    How to answer

    • Define knowledge representation and its role
    • Explain approaches: logic, semantic nets, frames, scripts, production rules
    • Discuss issues: adequacy, expressiveness, reasoning efficiency
    • Give examples for each approach
    • Conclude with desirable properties
13. 7 Marks
    aWhat is control strategy? What are the characteristics of a good control strategy?
    How to answer

    • Define control strategy in search
    • List characteristics: systematic, efficient, flexible, optimal
    • Explain forward vs backward, depth vs breadth control
    • Give role in deciding next state / rule to apply
    • Example of good strategy
14. 7 Marks
    bIf two fuzzy sets are $A = \{1/2 + 0.5/3 + 0.3/4 + 0.2/5\}$ and $B = \{0.5/2 + 0.7/3 + 0.2/4 + 0.4/5\}$. Find

    1. i)$A \cup B$.
    2. ii)$A \cap B$
    3. iii)$A'$
    4. iv)$B'$
    How to answer

    • Recall fuzzy union = max, intersection = min, complement = 1-mu
    • Compute membership for elements 2,3,4,5
    • List results for $A \cup B$ and $A \cap B$
    • List results for $A'$ and $B'$
    • Present in fuzzy set notation
15. 14 Marks
    Write short notes on following (any two)

    1. i)Monotonic and Non-monotonic Reasoning
    2. ii)NLP phases and fundamental problems
    3. iii)Script
    How to answer

    • Define monotonic vs non-monotonic / NLP phases / script
    • Give key characteristics and examples
    • Explain importance in AI reasoning / language / understanding
    • Compare / list steps with diagram if needed
    • Conclude with applications