Bharathiar University 2008 M.Phil Computer Science - Neural networks - Question Paper

M.Phil. Computer Science - Neural networks May 2008
Time : 3 hoursMaximum : 100 marks
ans any 5 ques..
All ques. carry equal marks.
1. Describe briefly about any 5 learning methods in Neural Networks.
2. Illustrate the step by step procedure of the back propagation algorithm.
3. What are the 4 classifications of associative memory architectures? discuss.
4. How the BAM works? discuss using BAM energy function.
5. Cluster all the five bit binary vectors other than the all zero vector using ART one algorithm. Study the effect of vigilance parameter on the resulting clusters.
6. Let and

,
.
Determine the implication relations
(a) IF is THEN is
(b) IF is THEN is .
7. What are the 2 classifications of CRISP Logic? discuss.
8. Bring out the necessity of fuzzy databases and discuss.

Earning:   Approval pending.