Jawaharlal Nehru Technological University Kakinada 2010-1st Sem M.Tech Computer Science & Engg. - Information Security SUBJECT CODE: C4004 REGULAR S DATA MINING AND KNOWLEDGE DISCOVERY

1. a. Write the purpose of feature subset selection and explain the related techniques.
b. explain the suitability of different proximity measures to objects involving various kinds
of attributes.

2. a. explain the greedy algorithm for Decision Tree induction.
b. explain different approaches for improving the generalization accuracy of a
classifier.

3. a. Write the characteristics of closest neighbor classifiers.
b. Write the significance of kernel trick in non – linear SVM model development.

4. a. explain different interestingness measures used in Association Mining.
b. explain how FP – growth algorithm overcomes the limitations of A Priori
algorithm.

5. a. What is meant by sequential trend discovery? Illustrate with an example.
b. Write an algorithm for sequential trend discovery.

6. a. explain K- Means algorithm for partitional clustering.
b. explain how the initial Centroids can be chosen for K- means algorithm.

7. a. explain different measures for cluster Validity.
b. Write the significance of Silhouette Coefficient for clustering of data objects.

8. a. What is a search engine? explain its architecture.

b. Write an algorithm for page ranking and explain how it can be used by a search

engine.

Earning:   Approval pending.