Reg No.:_ Name:

FIFTH SEMESTER B.TECH DEGREE EXAMINATION(R&S), DECEMBER 2019

Max. Marks: 100

1

2

4

5 a)
b)

6 a)
b)

7 ஐ
b)

8

9

1:192135 Pages:3

APJ ABDUL KALAM TECHNOLOGICAL UNIVERSITY

Course Code: CS361
Course Name: SOFT COMPUTING

PARTA
Answer all questions, each carries 3 marks.
Compare and contrast biological neuron and artificial neuron (3 points)
Obtain the output of the neuron for a network with inputs are given as [x1, x2] =
[0.7, 0.8] and the weights are [w1, w2] = [0.2, 0.3] with bias = 0.9.
Use i) Binary sigmoidal activation function
ii) Bipolar sigmoid activation function
State the training algorithm for multiple output classes in Perceptron.
What is the role of Widrow-Hoff rule in Adaptive Linear neuron? Give
appropriate equations.
PART ‏تا‎
‎Answer any two full questions, each carries 9 marks.
List any four activation functions with their equations and graphs.
Implement NOR(x,,x,) where x,,x, ಆ [0,1] using MP neuron.
Draw the flowchart of Hebb training algorithm.
Design a Hebb net to implement NOR function using with bipolar inputs and
targets.

Find the weights required to perform the following classifications using
perceptron network:

The vectors (1, 1, -1, -1) and (1,-1. 1, -1) are belonging to a class having target
value 1. The vectors (-1, -1, -1, 1) and (-1, -1, 1, 1) are belonging to a class
having target value -1. Assume learning rate 1 and initial weights as 0.

Draw the architecture of Back-Propagation network. Write its testing algorithm.

PART ட்‌
Answer all questions, each carries 3 marks.
Why the Law of Excluded Middle does not get satisfied in fuzzy sets?
Consider a local area network (LAN) of interconnected workstations that
communicate using Ethernet protocols at a maximum rate of 20 Mbit/s. The two

fuzzy sets given below represent the loading of the LAN:

Page 1 of 3

Duration: 3 Hours

Marks

(3)
(3)

(3)
(3)

(4)
(5)
(4)
(5)

(5)

(4)

(3)
(3)