NR
Code No: 75228/MT
M.Tech. – II Semester Regular Examinations, September, 2008 PATTERN RECOGNITION (Neural Networks) Time: 3hours
1.a) b)
Max. Marks:60
Answer any FIVE questions All questions carry equal marks --What are the components of pattern recognition system? Explain clearly each one with an example. Distinguish between supervised and unsupervised learning methods.
2.
Define the following terms: (i) Feature space (ii) Loss function (iii) Decision rule (iv) Bayes risk (v) Class – conditional probability density function (vi) Likelyhood ratio.
3.
Explain clearly about univariate and multivariate normal density functions.
4.
Explain the maximum-likelyhood estimation method for Gaussian case.
5.
What are the different criterian functions for clusstering? Explain each one.
6.
Write notes on (i) Principle component analysis (ii) Non – linear component analysis.
7.
Define discrete line markov process. What are the three basic problems for HMMs? Explain.
8.
Distinguish between continuous and discrete HMMs. What are the various types of continuous HMMs? Explain briefly.
&_&_&_&