Annamalai University 2004 Diploma Security Management m.phil-cs:neural networks and fuzzy systems - Question Paper

M.Phil. DEGREE exam – May 2004
(COMPUTER SCIENCE)

(PAPER - III)
132 - Neural Networks and Fuzzy Systems

Time: three Hours
Maximum: 100 Marks

PART - A (5 x eight = 40)

ans ALL ques..
All ques. carry equal marks.

1. (a) What is a neural network? explain the basic concepts of neural networks,

(OR)

(b) What is delta rule ? -Discuss.

2. (a) What do you mean by' simulated annealing ? -Discuss.

(OR)

(b) discuss an algorithm of generalized Hop - filed network.

3. (a) Write four short note on counter propagation networks.

(OR)

(b) With a neat diagram provide a general description of ART network.

4. (a) What is cognitron ? -Discuss.

(OR)

(b) explain neo-cognitron.

5. (a) Compare Fuzzy and crisp-sets.

(OR)

(b) List the applications of Fuzzy Logic. explain any 1 in detail.

PART - B (3 x 20 = 60)

ans any 3 ques..
All ques. carry equal marks.

6. What is learning? What do you mean by supervised and unsupervised learning? discuss back propagation algorithm in detail.

7. (a) Write a short note on Boltzmann machines.

(b) What do you understand by BAM ? - explain.

8. (a) Draw a flow - chart for the ART-z training mechanism.

(b) With a neat diagram explain ART - I model.

9. What do you mean by optical neural networks? explain the architecture and application of optical neural networks.

10. (a) Write a short note on Fuzzy relations.

(b) explain the application of Fuzzy logic in trend recognition.

Earning:   Approval pending.