M.Sc. DEGREE EXAMINATION, NOVEMBER/DECEMBER 2017.
Third Semester
Computer Science
Paper II DATAMINING
2 20932 AA
Time 3 Hours Max. Marks 70
SECTION — A
Answer any FIVE questions. 6 30 Marks)
Each question carries 6 marks.
1. Describe the various data mining primitives.
2. List out major differences between the star schema and the snowflake schema.
3. What is meant by data discretization? Explain about the entropy based
discretization.
4. State and describe different components of data warehouse? How the data
warehousing differ from relational data model.
5. Write about Naïve Bayesian classification.
6. State the criteria for evaluation of classification and prediction methods.
7. State the requirements of clustering in data mining.
8. What is outlier? List out different types of outliers.
SECTION — B
Answer FOUR questions. 10 40 Marks)
Each question carries 10 marks.
UNIT I
9. Discuss about major issue in typical data mining system.
Or
10. What is meant by missing values in data cleaning? How to fill these missing
values?

UNIT II
11. What is data transformation? Describe the various data transformation techniques.
Or
12. Explain multidimensional data model with suitable example.
UNIT III
13. A database has five transactions. Let min _sup 60% and min_conf 80%
Tid Items Bought
T1 Y
T2 Y
T3 E
T4 Y
T5 E
Use Apriori to find frequent item sets of first four transactions. Mark non
frequent items.
Or
14. Explain about decision tree induction with suitable example.
UNIT IV
15. Briefly explain the feature of the following clustering algorithm:
BIRCH DBSCAN.
Or
16. What is Hierarchical method of clustering? Differentiate Agglomerative and
Divisive Hierarchical Clustering.
———————