SCCE B.Tech Old Question Paper 07A81205 Pattern Recognition : Sree Chaitanya College Of Engineering

Name of the College : Sree Chaitanya College Of Engineering
University : JNTUH
Department : Information Technology
Subject Code/Name : 07A81205/PATTERN RECOGNITION
Year/Sem : IV/II
Website : http://scce.ac.in/web/
Document Type : Old Question Paper

Download Model/Sample Question Paper : https://www.pdfquestion.in/uploads/scce.ac.in/4964-07A81205-PATTERNRECOGNITION.pdf

SSCE Pattern Recognition Old Question Paper

Code No: 07A81205
IV B.Tech II Semester Examinations,April/May 2012

Related : Sree Chaitanya College Of Engineering 07A71201 Information Retrieval Systems B.Tech Question Paper : www.pdfquestion.in/4963.html

Information Technology
Time: 3 hours
Max Marks: 80
Answer any FIVE Questions :

Set No. 1

All Questions carry equal marks :
1. For each of the following datasets, construct a normal plot, and decide if the data appear to be approximately normally distributed.
(a) 35, 43, 46, 48, 51, 55, 58, 65
(b) 2.0, 3.0, 3.2, 3.5, 3.7, 3.9, 4.0, 4.2, 4.4, 4.4, 4.5, 4.8, 5.0, 5.1, 5.4, 5.8, 6.1 [8+8]

2. Explain the Bayesian estimation or Bayesian learning approach to pattern classification problems. [16]
3. Discuss the state transition matrix and state-transition coecients for 4-state left-right Model. [16]

4. Class A has a symmetric triangular density ranging from 0 to 4, and class B has a uniform density ranging from 2 to 6. The prior probabilities and costs are the same for both classes.
(a) Where are the optimal decision regions?
(b) What are the probabilities of error for class A and for class B if these decision regions are used. [16]

5. Write short notes on the following:
(a) Applications of normal mixtures in unsupervised learning
(b) Mixture density
(c) Component densities
(d) Mixing parameters. [16]

6. (a) Given the observation sequence O=(o1,o2, …… oT ) and the model =
(A,B,) how do we choose a corresponding state sequence q=(q1,q2,…….qT ) that is optimal in some sense (i.e. best explains the observations)?
(b) Explain N-state urn-and-ball model. [8+8]

7. Consider the use of multidimensional scaling for representing the points x1 = (1, 0)t, x2 = (0, 0)t, and x3 = (0, 0)t, in one dimensions. To obtain a unique solution, assume that the image points satisfy 0 = y1< y2< y3:
Show that the criterion function Jee is minimized by the configuration with y2 = (1 + p 2 )/3 and y3 = 2y2. [16]

8. Distinguish between the preprocessing, feature extraction and classification operations of pattern recognition system. [16]

Set No 2

1. Explain the class- conditional densities in Bayesian estimation. [16]
2. (a) How do we adjust the model parameters =(A, B, ) to maximize P(O/)?
(b) Explain the discrete-time Markov process. [8+8]

3. Explain the related minimum variance criteria in clustering with examples. [16]
4. Explain the functional structure of a general statistical Pattern classifier with neat diagram. [16]

5. (a) Find the mean and variance of a standard normal distribution.
(b) Explain decision regions for two-dimensional Gaussian data. [8+8]

6. Explain non-linear component analysis with neat diagram. [16]
7. Explain about error rate, risk multiplier classifiers of Post processing in pattern recognition system. [16]

8. (a) In which case Hidden Markov model parameter set to zero initially will remain at zero throughout the re-estimation procedure.
(b) Constraints of the left-right model have no effect on the re-estimation procedure. Justify. [8+8]

IV B.Tech II Semester Examinations,April/May 2012 :
Pattern Recognition :
1. (a) Write sum-of sqared functions for multidimensional scaling.
(b) How do you compute the gradients of criterion function of multidimensional scaling? [8+8]

2. (a) Explain the concept of classication in pattern recognition system with examples.
(b) Explain the concept of post processing in pattern recognition system with examples. [8+8]

3. (a) Explain the marginal density functions
4. (a) Explain the general principle of maximum likelihood estimation.
(b) Find the maximum likelihood estimate for in a normal distribution. [8+8]

5. (a) Write the re-estimation formulas for the coefficients of the mixture density.
(b) Discuss the state transition matrix for 4-state ergodic model and 6-state parallel path left- right model with examples.

6. Some data with features x and y (see the following table) were randomly selected from a population that consists of classes A and B. What is the probability that a new sample with x = 0, y = 1 belongs to class A? Make only necessary assumptions and list them.

7. (a) Explain the concept of decision boundary in design of simple classifiers.
(b) Explain the design cycle of pattern recognition system and also explain the computational complexity in the design. [8+8]

8. What are the restrictions placed on the form of the probability density function to ensure that the parameters of the pdf can be re-estimated in a consistent way?