b)

b)

b)

b)

b)

0)

0)

03000EC360052102

PART B
Answer any two full questions, each carries 15 marks
Explain fuzzy rule based system with the help of block diagram.
With proper examples, compare supervised and unsupervised learning.
Implement XOR function using McCulloch-Pitts neuron (take binary data).
Explain the various defuzzification methods.
Let X = {a, 0, ०, d} Y= {1,2,3,4} and A= {(a,0), (0,0.6), (c,0.8), (6,1))
B= {(1,0.2), (2,1), (3,0.8),}; C= ((1,0), (2,0.4), (3,1), (4,0.8)}
Determine the implication relation
1. If X is A THEN Y is B else Y is C
2. If Xis A THEN Y is 8
Give the mathematical representation of an artificial neural network. Construct a
feed-forward network with four input nodes, two hidden nodes and three output
nodes.
PART C
Answer any two full questions, each carries 20 marks
Explain various stopping conditions for genetic algorithm flow.

With a neat diagram, explain the back propagation algorithm for training MLP.
Briefly explain the various encoding techniques used in genetic algorithm.

If the activation function of all hidden unit is linear, then show that MLP is
equivalent to single layer perceptron.

Explain the concept of linear separability.

Implement OR logical function with binary inputs & bipolar outputs using
perceptron training algorithm.

Explain various selection strategy techniques in Genetic Algorithm.

मे मर ‏بد بد‎

Page 2 of 2

(9)
(6)
(7)
(8)
(8)

(7)

(10)
(10)
(10)
(5)

(5)
(10)

(10)