(SEM V) THEORY EXAMINATION 2022-23 NEURAL NETWORKS & FUZZY SYSTEM

Course: B.Tech (Semester V)

Subject: Neural Networks & Fuzzy System

Subject Code: KEE-056

Time: 3 Hours

Total Marks: 100

Instructions: Attempt all sections. If data is missing, assume suitable values.

Section A – Short Answer Questions (2 × 10 = 20 Marks)

Answer all questions briefly:

Applications of artificial neural networks.

Draw a 2–4–2 feedforward neural network.

List various tuning parameters in backpropagation.

Compute neuron output when net input = 0.54 with binary sigmoidal activation.

Difference between fuzzy and crisp sets.

Explain linguistic variables.

Define universal set in fuzzy set theory with example.

Describe a fuzzy inference system.

Describe strong α-cut fuzzy set.

Define L–R type fuzzy numbers.

Section B – Descriptive Questions (10 × 3 = 30 Marks)

Attempt any three:

Compare and contrast biological and artificial neurons with diagrams.

Discuss the importance of the sigmoidal function in backpropagation. Derive the weight adjustment equation ΔW = η{OH}.

Explain components of a fuzzy logic control system with block diagram.

For given fuzzy sets MP (medium power) and HP (high power), demonstrate union, intersection, complement, and difference operations.

Explain the structure of a neural expert system in detail.

Section C – Long Answer Questions (10 × 5 = 50 Marks)

Q3.
(a) Analyze the necessity of activation functions and explain their types, or
(b) Train a hetero-associative memory network using Hebb’s rule for given input–output vector pairs.

Q4.
(a) Explain step-by-step backpropagation learning algorithm, or
(b) For an input neuron with input = 2, weight = 2.3, and bias = –3, find neuron output for:

Unit step (threshold = 1)

Linear (slope m = 2)

Bipolar sigmoidal (λ = 0.5)

Q5.
(a) Explain the meaning of defuzzification and its different methods, or
(b) Given fuzzy sets of speed and fuel consumption for a rocket, determine relation R~\tilde{R}R~ between them.

Q6.
(a) Design a fuzzy logic controller for an air conditioner (inputs, outputs, and rule base), or
(b) Explain the role of fuzzy logic in healthcare systems with a real-life example.

Q7.
(a) Explain the synergy between neural and fuzzy systems and their key characteristics, or
(b) Analyze the different steps of fuzzy backpropagation training.