Download as doc, pdf, or txt
Download as doc, pdf, or txt
You are on page 1of 2

BANNARI AMMAN INSTITUTE OF TECHNOLOGY

SATHYAMANGALAM
Ph.D DEGREE EXAMINATION APRIL/MAY 2012
10SE61 PATTERN RECOGNITATION
Time: 3 hours

Maximum marks: 100


Answer All Questions

Part - A

(10 x 2 = 20 marks)

1. Define discriminate functions.


2. What is meant by supervised learning?
3. Define unsupervised classification.
4. How compact classifier is used in four class problem?
5. What are the elements of formal grammars?
6. Define parsing.
7. What is meant by entropy minimization?
8. Compare the biological neuron and artificial neuron.
9. Define unsupervised learning.
10. How the patterns are classified using Genetic Algorithm?

Part - B

(5 x 16 = 80 marks)

11. (a) Discuss in detail about Maximum likelihood estimation.

(16)

OR
(b) Explain in detail about LMSE algorithm

(16)

12. (a) Explain in detail about clustering concepts and C-means algorithm (16)

OR
(b) Briefly explain the graph theoretic approach to pattern clustering. (16)

13. (a) Explain in detail about string generation as pattern description.

(16)

OR
(b) Discuss in detail about stochastic grammars and applications.
14. (a) Explain in detail about karhunen Loeve transformation.

(16)
(16)

OR
(b) Discuss in detail about Binary feature selection.

(16)

15. (a) Discuss in detail about neural network structure for pattern Recognition.
(16)
OR
(b) Explain in detail about fuzzy logic and fuzzy pattern classifiers

(16)

You might also like