Data Warehousing and Data Mining 2074

                                    Group - A

Attempt any two Questions                     (10 x 2 = 20)

1.  You are given the transaction data shown below from a fast food restaurant. There are 9 distinct transactions (order 1 to order 9). There are total 5 meal (M1 to M5) involved in transactions.

Minimum support =2, Minimum confidence = 0,7

Apply the Apriori algorithm to the database to identify frequent k-itemset and find all strong association rules.

2.  Why do we need to preprocess the data before running the algorithm? What are the processes for this? Explain. Give some examples of noise that must be removed in data while extracting the pattern.

3.  List the two steps used in classification approach with its issues. Is this right decision to use neural network always as a classifier? Give your opinion. Discuss the working mechanism of back propagation classification algorithm.

                                     Group - B

Attempt any eight Questions                     (8 x 5 = 40)

4. List and describe the five primitives for specifying a data mining task.

5.  Describe the types of data used in data mining.

6.  Explain the similarities and dissimilarities between operational database and data warehouse.

7.  List the types of OLAP operations with example.

9.  Why data cube computation is essential task in data mining? Describe general strategy in data cube computation.

10.  Describe the different components of a data warehouse.

11.  Define dimension table and fact table. What makes the necessity of multidimensional data model?

12.  Discuss the approach behind Bayesian classification. Why smoothing technique is necessary in Bayesian classification?

13.  Write short notes on (any two):

            a)  Concept hierarchy

            b)  Data mining Query Language

            c)  Text mining

            d)  ROLAP vs MOLAP