UNIV4VINU | ENGINEERING ACHIEVERS

CS 1004 – DATA WAREHOUSING AND MINING, APRIL / MAY 2008

PART A – (10 x 2 = 20 marks)

1. Compare OLTP and OLAP systems.

2. What is Data Warehouse Metadata?

3. What is Dimensionality Reduction?

4. What is Concept Description?

5. List two interesting measures for association rules.

6. What are Iceberg queries?

7. What is classification?

8. What is cluster analysis?

9. What is Web Usage Mining?

10. What is Visual Data Mining?

PART B – (5 x 16 = 80 marks)

11. (a) Briefly compare the following concepts. Explain your points with an example

(i) Snowflake schema, fact constellation, star net query model [Marks 5]

(ii) Data cleaning, data transformation, refresh [Marks 5]

(iii) Discovery-driven cube, multifeature cube, virtual warehouse [Marks 6]

(b) What are the difference between three main types of data usage: information

processing, analytical processing and data mining? Discuss the motivation behind OLAP

mining. [Marks 16]12. (a) For class characterization, what are the main differences

between a data cube based implementation and a relational implementation such as

attribute-oriented induction. Discuss which method is most efficient and under what

condition this is so. [Marks 16]

Or

(b) (i) List and discuss the various data mining primitives. [Marks 8]

(ii) With relevant examples discuss the role of statistics in data mining. [Marks 8]

13. (a) Explain with an algorithm, how to mine single dimensional Boolean Association

Rules from transactional database. Give relevant example. [Marks 16]

Or

(b) With an algorithm explain constraint-based association mining. Give relevant example.

[Marks 16]

14. (a) What are Bayesian classifiers? Explain in detail about:

(i) Naïve Bayesian classification [Marks 8]

(ii) Linear and multiple regression. [Marks 8]

Or(b)

Why is outline mining important? Briefly describe the different approaches behind