You are on page 1of 3

Reg. No.

Question Paper Code : 41008

B.E./B.Tech. DEGREE EXAMINATIONS, NOVEMBER/DECEMBER 2021.

Fifth Semester

Computer Science and Engineering

OEC 552 – SOFT COMPUTING

(Common to: Information Technology)

(Regulations 2017)

Time : Three hours Maximum : 100 marks

Answer ALL questions.

PART A — (10 × 2 = 20 marks)

1. Differentiate fuzzy set and crisp set.

2. Consider two fuzzy sets A and B defined as follows

~
A = {(1, 0, 1), (2, 0, 5), (3,0,8 ), (4,1,0 ), (5, 0, 7 ), (6, 0, 2)}

~
B = {(1,1, 0 ), (2, 0, 8 ), (3, 0, 4 ), (4, 0, 1)} Find A ∪ B .

3. What is a biological neuron?

4. Differentiate supervised and unsupervised learning.

5. What is defuzzification?

6. What is a fuzzy expert system?

7. What is a genetic algorithm?

8. Define optimization.

9. Present of outline of genetic neuro hybrid system?

10. What are hybrid fuzzy controllers?


PART B — (5 × 13 = 65 marks)

11. (a) Outline the evolution of artificial neural networks. (13)

Or

(b) (i) Outline tolerance and equivalent relations with an example. (5)

(ii) Explain Cartesian product in fuzzy relations with an example. (8)

12. (a) Explain the characteristics and architecture of single and multi layered
feed forward neural network with a diagram. (13)

Or

(b) Explain Hopfield network with a diagram. (13)

13. (a) What is a fuzzy membership function? Outline the types of fuzzy
membership function with an example and a diagram.

Or

(b) Elaborate fuzzy reasoning with examples.

14. (a) Outline the genetic algorithm operators with an example. (13)

Or

(b) Use genetic algorithm to maximize f (x ) = x 2 + 20 Let the initial


Population be the binary equivalent of number 8, 5, 9, and 4 with four
bits. Find the average fitness and the member with high fitness value.
Use one point cross over at site 2 with the high Fitness chromosome. Use
both Roulette wheel selection and Rank selection to generate the
offspring for the next generation. (13)

15. (a) Explain how solution to travelling salesman problem is achieved using
genetic algorithm with an example. (13)

Or

(b) Outline the building blocks of a genetic fuzzy systems with a diagram.(13)

2 41008
PART C — (1 × 15 = 15 marks)

16. (a) Write a note on back propagation network. Perform single iteration and
find the error rate for the given network which has two inputs I1 and I2
and a single output O. (15)
0.1
0.4 0.2
0.4 0.1
– 0.2
– 0.5
– 0.7
0.2
Or
(b) Explain Kohonens algorithm and consider a Kohonen net with two
clusters and 3 input units. The weight vector for cluster units are (0.9,
0.7, 0.6) and (0.4, 0.3, 0.5). Find winning cluster for input vector 0.4, 0.2
and 0.1. Use learning rate 0.2. Find the new weights for the winning
unit. (15)

—————————

3 41008

You might also like