(SEM VII) THEORY EXAMINATION 2024-25 ARTIFICIAL INTELLIGENCE

B.Tech Engineering 0 downloads

₹29.00

ARTIFICIAL INTELLIGENCE (KCS071)

B.Tech – Semester VII

Time: 3 Hours | Max Marks: 100

SECTION – A (10 × 2 = 20 Marks)

(Attempt all questions in brief)

(a) Define Artificial Intelligence.

Artificial Intelligence (AI) is a branch of computer science that deals with creating machines or systems capable of performing tasks that normally require human intelligence such as learning, reasoning, problem-solving, and decision-making.

(b) Name two most common AI problems.

Two common AI problems are search problems and knowledge representation problems.

(c) Explain problem-solving methods.

Problem-solving methods in AI involve representing a problem as a state space and using search techniques such as uninformed and informed search to find a solution.

(d) Recall local search.

Local search is a search technique that starts with an initial solution and improves it iteratively by exploring neighboring states without considering the entire state space.

(e) Explain forward chaining in inference systems.

Forward chaining is a data-driven inference method where reasoning starts from known facts and applies rules to derive new facts until a goal is reached.

(f) What is ontological engineering?

Ontological engineering is the process of designing and developing formal representations of knowledge using concepts, relationships, and rules within a specific domain.

(g) Explain the main components of an architecture for intelligent agents.

An intelligent agent architecture consists of sensors, actuators, environment, perception module, decision-making module, and learning component.

(h) Evaluate the purpose of negotiation and bargaining in multi-agent systems.

Negotiation and bargaining help agents resolve conflicts, allocate resources efficiently, and reach mutually beneficial agreements in a multi-agent environment.

(i) Name two applications of AI.

Two applications of AI are speech recognition and expert systems.

(j) Distinguish Information Retrieval from Information Extraction.

Information Retrieval focuses on finding relevant documents, whereas Information Extraction extracts specific structured information from unstructured data.

SECTION – B (Attempt any THREE) (3 × 10 = 30 Marks)

2(a) Discuss the impact of quantum computing on intelligent agents in AI.

Quantum computing enhances AI by enabling faster computation, parallelism, and efficient optimization. Intelligent agents can process large datasets quickly, improve learning efficiency, and solve complex problems that are infeasible for classical computers.

2(b) How do heuristics contribute to problem-solving in AI?

Heuristics guide search algorithms by estimating the cost to reach the goal, reducing search space and time. Common techniques include hill climbing, best-first search, and A* algorithm.

2(c) Examine unification in Prolog and its significance.

Unification is the process of matching variables with constants or other variables. It is essential for logical inference in Prolog, enabling rule application and pattern matching.

2(d) Appraise reputation management in multi-agent systems.

Reputation management allows agents to evaluate trustworthiness based on past behavior. It improves cooperation, reduces fraud, and enhances decision-making in distributed systems.

2(e) Illustrate real-world applications of AI and their impact.

AI is used in healthcare (diagnosis), finance (fraud detection), transportation (self-driving cars), education (personalized learning), and manufacturing (automation), improving efficiency and accuracy.

SECTION – C (Attempt ANY ONE) (1 × 10 = 10 Marks)

3(a) Differentiate between reactive and proactive intelligent agents with applications.

Reactive agents respond directly to environmental changes without planning. Proactive agents plan ahead, set goals, and take initiative.

Reactive agents are used in robotics, while proactive agents are used in personal assistants and autonomous systems.

3(b) How can future AI address limitations in Natural Language Processing?

Future AI can overcome NLP limitations using deep learning, contextual understanding, multilingual models, and improved semantic representation, enabling better language comprehension.

SECTION – D (Attempt ANY ONE) (1 × 10 = 10 Marks)

4(a) Discuss primary problem-solving methods in AI and differences from traditional computing.

AI problem-solving uses heuristic search, learning, and reasoning, whereas traditional computing follows fixed algorithms. AI handles uncertainty and adaptability better.

4(b) Explain backtracking search in CSPs with advantages and limitations.

Backtracking systematically explores solutions and backtracks when constraints are violated. It is simple but inefficient for large CSPs.

SECTION – E (Attempt ANY ONE) (1 × 10 = 10 Marks)

5(a) Evaluate First Order Predicate Logic (FOPL) and compare with propositional logic.

FOPL is more expressive, allowing quantifiers and predicates, while propositional logic uses simple true/false statements. FOPL handles complex reasoning but is computationally complex.

5(b) Explain how reasoning systems handle mental events and objects.

Reasoning systems model beliefs, intentions, and goals using symbolic representations, enabling intelligent decision-making in AI applications.

SECTION – F (Attempt ANY ONE) (1 × 10 = 10 Marks)

6(a) Explain negotiation and bargaining in multi-agent systems.

Agents communicate preferences, exchange offers, and apply strategies to reach agreements while maximizing individual utility.

6(b) Discuss challenges in establishing trust among agents.

Challenges include lack of transparency, malicious behavior, dynamic environments, and incomplete information.

SECTION – G (Attempt ANY ONE) (1 × 10 = 10 Marks)

7(a) Discuss steps involved in speech recognition and its importance.

Speech recognition involves signal processing, feature extraction, pattern matching, and language modeling. It enables natural human-computer interaction.