00000EC467121902

Pages: 2

Reg No.: Name:

Max. Marks: 100

1 a)
b)
2 a)
b)
3 a)
b)
4 a)
b)
5 a)
b)
6 a)
b)
7 a)
b)

APJ ABDUL KALAM TECHNOLOGICAL UNIVERSITY

Seventh semester B.Tech examinations (S), September 2020

Course Code: EC467
Course Name: PATTERN RECOGNITION

PARTA

Answer any two full questions, each carries 15 marks.
Explain the design cycle of a pattern recognition system.
Prove that a Bayes classifier is equivalent to a minimum distance classifier,
assuming that the feature vector is Gaussian.
Compare supervised, unsupervised and reinforcement learning techniques.
Explain the principal component analysis for dimensionality reduction.
Explain Hidden Markov model and its role in the classifier design.

What is meant by the curse of dimensionality?

PART B
Answer any two full questions, each carries 15 marks.

Explain the Parzen window method for density estimation.

What is pruning in decision tree construction? Explain its significance.

Explain the Perceptron algorithm.

Explain the nonparametric methods for density estimation.

Formulate SVM as an optimization problem. How support vector machines can
be used for classification of data which are not linearly separable?

Illustrate the concept of a decision tree with the help of an example.

PART C
Answer any two full questions, each carries 20 marks.
Explain the boosting approach in classifier ensembles. Give details of Adaboost

algorithm.
What is X-OR problem in classification? With a neat diagram explain the

solution of XOR problem.

Page lof 2

Duration: 3 Hours

Marks
(5)
(10)

(6)
(9)
(10)
(5)

(9)
(6)
(10)
(5)
(10)

(5)

(10)

(10)