Papers / AKTU / B.Tech / Artificial Intelligence
AKTU · B.Tech · 5th semester

AKTU Artificial Intelligence (BCAI501) Question Paper 2024-25

UniversityDr. A.P.J. Abdul Kalam Technical University (AKTU), Lucknow
CourseB.Tech, Semester V
SubjectArtificial Intelligence (BCAI501)
ExamTheory Examination 2024-25 (odd semester, held January 2025)
Time · Marks3 hours · 70 marks
SourceAKTU question paper, via Notes Gallery
View original PDF See important topics

Note: Attempt all sections. In case of any missing data, choose suitably.

Section A

Q1. Attempt all questions in brief · 2 × 7 = 14 marks

  1. Define Artificial Intelligence. [CO1]
  2. Define "problem-solving" in AI. [CO1]
  3. Define problem-solving in the context of AI. [CO2]
  4. What is a Constraint Satisfaction Problem (CSP)? [CO2]
  5. What is First Order Predicate Logic (FOPL)? [CO3]
  6. Define an intelligent agent. [CO4]
  7. Explain the concept of machine translation. [CO5]

Parts (b) and (c) are printed this way on the original paper.

Section B

Q2. Attempt any three · 7 × 3 = 21 marks

  1. Compare and contrast different types of Intelligent Agents (Simple Reflex, Model-Based, Goal-Based, and Utility-Based). [CO1]
  2. Explain the difference between uninformed and informed search strategies with examples. [CO2]
  3. Discuss different types of knowledge representation techniques in AI. [CO3]
  4. Describe the negotiation process in multi-agent systems with examples. [CO4]
  5. Discuss the process and significance of information retrieval in AI. [CO5]

Section C

Q3 to Q7. Attempt any one part of each question · 7 × 5 = 35 marks

Q3

  1. Explain the evolution of Artificial Intelligence from symbolic AI to modern machine learning. [CO1]
  2. Explain the role of AI in enhancing decision-making processes in business applications. [CO1]

Q4

  1. Discuss the structure and components of a Constraint Satisfaction Problem (CSP). [CO2]
  2. Describe the process of solving a Sudoku puzzle using CSP methods. [CO2]

Q5

  1. Compare semantic networks and frames as knowledge representation methods. [CO3]
  2. Explain how reasoning systems are designed for decision-making in AI. [CO3]

Q6

  1. Explain how argumentation is used for resolving conflicts in multi-agent systems. [CO4]
  2. Explain how autonomous agents manage decision-making and execution. [CO4]

Q7

  1. Explain the steps involved in extracting structured information from unstructured data. [CO5]
  2. Explain how tokenization impacts natural language processing tasks. [CO5]

Most repeated topics

How many of the last 3 AKTU 5th-semester Artificial Intelligence papers (KAI501 2022-23 and 2023-24, BCAI501 2024-25) asked about each topic.

Intelligent agents and their types
3 / 3
Search strategies (A*, uniform-cost, informed vs uninformed)
3 / 3
First order predicate logic
3 / 3
Knowledge representation and ontology
3 / 3
Multi-agent systems (trust, negotiation, argumentation)
3 / 3
NLP, machine translation, information retrieval
3 / 3
Game playing and alpha-beta pruning
2 / 3
Constraint satisfaction problems
2 / 3
Computer vision and pattern recognition
1 / 3

Counted by LearnAIFast from the 2022-23, 2023-24 and 2024-25 odd-semester theory papers.

Answer hints

Q1(d). What is a CSP?

A Constraint Satisfaction Problem has a set of variables, a domain of allowed values for each, and constraints on which combinations are allowed. A solution assigns every variable a value without breaking any constraint. Examples: map colouring, Sudoku, timetabling.

Q2(a). Types of intelligent agents

Simple reflex agents act only on the current percept using condition–action rules. Model-based agents keep an internal state of the world, so they handle partial observability. Goal-based agents choose actions that reach a goal, using search or planning. Utility-based agents choose the action with the highest expected utility when several paths reach the goal. Give one example for each, such as a thermostat, a robot vacuum, a route planner and a self-driving car.

Q2(b). Uninformed vs informed search

Uninformed (blind) search uses only the problem definition: BFS, DFS, uniform-cost, iterative deepening. Informed (heuristic) search uses an estimate h(n) of the distance to the goal: greedy best-first and A*, where f(n) = g(n) + h(n). A* is optimal when h(n) never overestimates. Compare them on completeness, optimality, time and memory.

Q4(b). Sudoku as a CSP

Variables are the 81 cells, the domain is 1–9, and the constraints are all-different in every row, column and 3×3 box. Solve with backtracking search plus constraint propagation (forward checking or arc consistency), and pick the cell with the fewest remaining values first (MRV heuristic).

Q7(b). Tokenization in NLP

Tokenization splits text into units (words, subwords or characters) that later steps work on. The choice affects vocabulary size, how unknown words are handled, and accuracy. Subword methods like BPE and WordPiece, used by transformer models, split rare words into known pieces. Mention the challenges: punctuation, contractions, and languages like Hindi with complex scripts or no spaces.

Other years

2023-24 (KAI501) · PDF 2022-23 (KAI-501) · PDF Machine Learning Techniques 2023-24 All AI & ML papers

This paper is shared for study purposes only. Copyright belongs to AKTU. If you own the copyright and want it removed, email learnwithskillnest@gmail.com and we will take it down.