USN 10IS74

(08 Marks)(07 Marks)(05 Marks)

(04 Marks)(08 Marks)

(08 Marks)

(10 Marks)(10 Marks)

(08 Marks)

(08 Marks)(04 Marks)

(08 Marks)(08 Marks)(04 Marks)

Seventh Semester B.E. Degree Examination, Dec.2013/Jan.2O14

Data Warehousing and Data MiningMax. Marks:100Time: 3 hrs.

"'1, ,

2a.b.c.

Note: Answer FIVE full questions, selectingat least TWO questions from each part.

()o(lFq

C)

o

oXbo-d=

GV-oo0ttrco

.= 6l

xbots.JFO

o>

6:

oO

-!

b0ccg(dE:-6

-o'Enor=

o. E-

o=ia tEctr!o5?. .=Xqco0o=o- ;jtr>v!o

U<:..i()oZ

oo.

PART _ AL a. lVhat is Operational Data Store (ODS)? Explain the ODS structure with a neat diagram.

b.

c.

What is ETL? Explain the steps in data cleaning.What are the guidelines for implementing data warehouse?

Distinguish be,tween OLTP and OLAP.Explain the operations of data cube with suitable example_g.,

Write short notes on: i) ROLAPiii).,Data cube

ii) MOLAP-.iv) FASMI characteristics

3 a. Discuss the tasks of data mining with suitable examples.b. Explain shortly any five data preprocessing approaches.

4 a. Develop the Apriori algorithm fo. ga;;*tirrg t.qr.rt item set.b. Consider the transaction data set: I i,

Tid I 2 J 4' 5 6 7 8 9 l0Items {a,b} {b,c,d} {a,c,d,e} {a.d,e} {a,hc,} {a,b,c,d} {al {a,b,c} {a,b,d} {b,c,e}

Construct the FP tree by showing the trees separBtely after reading each transactron.

f,4.b.c.

PART _ BExplain Hunt's algorithm and illustrate its working (08 Marks)What is Ruls:Based classifier? Explain how a Rule-Based classifier works. (08 Marks)Write the algorithm for K-nearest neighbor classification. (04 Marks)

a. What is Bayes theorem? Show how is it used for classification. ' (06 Marks)b. Discuss the methods for estimating predictive accuracy of classification method. (10 Marks)c. i,,.What are two approaches for extending the binary classifiers to handle multiclass problems?

(04 Marks)

a. List and explain four distance measu-res to compute the distance between a pair of points andfind out the distance between two objects represented by attribute values (I, 6.2" 5. 3) and

b.

c.

(3, 5, 2, 6, 6) by using any two of the distance measures.Explain the cluster analysis methods briefly.What are the features of cluster analysis?

Write short note on the following:Web content miningText miningSpatial databases miningMining temporal databases

a.

b.

c.

d. (20 Marks)

For More Question Papers Visit - www.pediawikiblog.com

For More Question Papers Visit - www.pediawikiblog.com

www.pediawikiblog.com