Andhra University 2005-2nd Year B.Tech Computer Science and Engineering Forth Semester - DATAWARE HOUSING AND DATA MINING (ELECTIVE-II) - Question Paper

MODEL PAPER

B. Tech (CSE) Degree exam

Forth Year - 2nd Semester

DATAWARE HOUSING AND DATA MINING (ELECTIVE-II)

Time: three hrs
Max Marks: 70

First ques. is Compulsory

ans any 4 from the remaining ques.

All ques. carry equal marks

ans all parts of any ques. at 1 place

1. Briefly explain.
a. kinds of OLAP
b. Correlation analysis for handling redundancy.
c. Discretization
d. Ice-berg query.
e. Constraint -based rule mining
f. Non-linear regression
g. Ordinal variables for cluster analysis.

2. a. What is data mining? Briefly define the components of a data mining system.
b. What types of trends can be identified in a data mining system?

3. a. Write the differences ranging from operational database and data warehouse.
b. Briefly define 3-tier Data warehouse architecture

4. a. Write various approaches to data transformation.
b. Propose an algorithm in pseudo-code for automatic generation of a concept hierarchy for categorical data based on the number of distinct values of attributes in the provided schema.

5. a. explain the essential features of a typical data musing query language like DMQL.
b. Consider association Rule below, which was mined from the learner database at Big University:
Major(X. "science") &rarr status (X,"undergrad").

Suppose that the number of students at the university (that is, the number of task-relevant data tuples) is 5000. that 56% of undergraduates at the university major in science, that 64% of the students are registered in programs leading to undergraduate degrees, and that 70% of the students are majoring in science
a. calculate the confidence and support of above rule
b. Consider Rule beneath
Major (X,"biology") ? status (X,"undergrad"). [17%,80%]
Suppose that 30’% of science students are majoring in biology. Would you consider Rule two to be novel with respect to rule1? discuss.

6. a. explain why attribute relevance analysis is needed and how it can be performed.
b. Outline a data cube-based incremental algorithm for mining analytical class comparisons.

7. Write the A priori algorithm for discovering frequent item sets for mining single-dimensional Boolean Association Rule and explain different approaches to impiove its efficiency.

8. a. explain the backpropagation algorithm for neural network-based classification of data.
b. elaborate the various categoiies of clustering methods?

Earning:   Approval pending.