13

14

15

16

17

18

19

20

a)

b)

a)
b)
a)
b)
a)

b)
a)

b)
a)

b)

a)
b)

03000CS362052002

State the cause for uncorrelated estimates of pose. How can this issue be
handled?
How is prior probability related to posterior probability? What role do they
play in decision making?
Outline how to perform verification in model based vision.

PART E

Answer any four full questions, each carries 10 marks.

Explain perceptron learning algorithm.

Describe the minimum mean squared error method for classification.

Explain the working of support vector machines with neat illustrations.

Explain the ID3 algorithm for classification.

What are proximity measures? State two properties of a dissimilarity measure.
Mention any two examples for dissimilarity measures, with equations.

Explain the various steps involved in clustering, with a suitable example.
Mention the ways in which neural networks can be used for pattern
recognition.

How are genetic algorithms used for pattern classification?

Write a short note on Classification And Regression Trees (CART). Explain
with an example.

How are linear discriminant based classifiers different from tree based
classifiers?

Explain the K-Means algorithm for clustering, with an example.

What are the different types of features that could be used for clustering?

Page 2 of 2

(4)

(5)

(9)

(5)
(5)
(5)
(5)
(5)

(5)
(5)

(5)
(7)

(3)

(6)
(4)