Download AKTU B-Tech 8th Sem 2017-18 NCS 080 Pattern Recognition Question Paper

Firstranker's choice

Printed Pages: 02

Paper Id: 11821

--- Content provided by‍ FirstRanker.com ---

FirstRanker.com

Sub Code: NCS-080

Roll No.

Time: 3 Hours

B TECH

--- Content provided by FirstRanker.com ---

(SEM. VIII) THEORY EXAMINATION 2017-18

PATTERN RECOGNITION

Attempt all Sections. If require any missing data; then choose suitably.

Total Marks: 100

SECTION A

--- Content provided by‌ FirstRanker.com ---

Attempt all questions in brief. 2 x 10 = 20

1. a. Define the law of total probability.
2. b. What do you mean by dimension reduction in pattern recognition?
3. c. Write the difference between supervised learning and unsupervised learning.
4. d. How do we evaluate the performance of a classifier?

--- Content provided by‍ FirstRanker.com ---

5. e. What is Hidden Markov Model (HMM)?
6. f. What is discriminant function?
7. g. Write short notes on Gaussian mixture model.
8. h. Discuss cluster validation.
9. i. Write K-means clustering algorithm.

--- Content provided by FirstRanker.com ---

10. j. Write the difference between clustering and classification.

SECTION B

Attempt any three of the following: 10 x 3 = 30

1. a. What do you mean by learning and adaptation? Explain the components of a learning system.
2. b. Explain the Chi-Square test and discuss their significance in pattern recognition with suitable example.

--- Content provided by‍ FirstRanker.com ---

3. c. Explain the concept of expectation maximization with the help of an algorithm.
4. d. How K-nearest neighbor (KNN) method works? Explain with KNN estimation and KNN rule.
5. e. Explain Naïve Bayes Classifier.

SECTION C

Attempt any one part of the following: 10 x 1 = 10

--- Content provided by‍ FirstRanker.com ---

1. (a) Consider a two-class problem where the classes are labeled pen drive and laptop. Suggest a set of features that could be used to discriminate between these two classes of objects.
2. (b) What is Baysian Decision Theory? Discuss two-class category Classification in details.

FirstRanker.com

Attempt any one part of the following: 10 x 1 = 10

1. (a) Illustrate statistical and syntactic pattern recognition (SPR) approach.

--- Content provided by⁠ FirstRanker.com ---

2. (b) Explain normal density function and discuss its significance in pattern recognition?

Attempt any one part of the following: 10 x 1 = 10

1. (a) Write short notes on following-
   1. i. Maximum likelihood estimation
   2. ii. Bayesian Estimation

--- Content provided by FirstRanker.com ---

2. What is Parzen window? Explain. Derive the conditions for (i) convergence of mean (ii) convergence of variance.

Attempt any one part of the following: 10 x 1 = 10

1. (a) What do you mean by Fuzzy Decision making? Also discuss the Fuzzy Classification using suitable example.
2. (b) Name the different methods of non-parameter estimation strategies. What are the main differences between them?

Attempt any one part of the following: 10 x 1 = 10

--- Content provided by FirstRanker.com ---

1. (a) What are different clustering techniques? Why is clustering important? What is an agglomerative clustering algorithm? Explain.
2. (b) Write short notes on
   1. i. Cluster validation
   2. ii. Criteria function for clustering

--- Content provided by​ FirstRanker.com ---