Pune University BE IT Information Retrieval Question Papers

B.E. (Information Technology) INFORMATION RETRIEVAL (2008 Pattern) (Sem. – II)

Time :3 Hours]                                                                           [Max. Marks :100

Instructions to the candidates:-

1)           Answers to the two sections should be written in separate books.

2)           Neat diagrams must be drawn wherever necessary.

3)           Pigures to the right indicate full marks.

4)           Assume suitable data, if necessary.

SECTION – I

QI) a) With the help of block diagram explain typical Information Retrieval System.      [8]

b)        Explain Measures of Association in detail.                                                    [4]

c)        What do you mean by Cluster. State and Explain Cluster Hypothesis. [6]

OR

Q2) a) What do you mean by Cluster. State and explain Cluster Hypothesis. [4]

b)        State Zipf’s Law, Explain Luhn’s Idea.                                                           [6]

c)         Explain Graph Theoretic approach for clustering and draw the cluster derived from the given similarity matrix where Threshold = 0.89 and [8]

 I 0.6 2 0.6 3 0.9 .8 4 0.9 .7 .7 5 0.9 .6 .6 .9 6 0.5 .5 .5 .9

Explain Inverted Index file. How it can be used in Information Retrieval. [8] Explain Vector Model in detail.                                                                                                                [8]

OR

Explain with an example organization of records in Multi-lists, state its Advantages over Inverted Files.                                                                                                                      [8]

Explain the different kinds of Search strategies.                                           [8]

Discuss the five key abstractions used to design or model Digital Libraries.          [10]

Explain Harmonic Mean and E measure.                                                       [6]

OR

Explain need for Single Value Summaries and various strategies in detail. [10]

Write short note on TREC.                                                                               [6]

SECTION – II

Describe MIMD architecture with respect to Parallel IR. How inverted file is used for MIMD. [10]

Define Ontology? Explain in detail reasons to develop Ontology? [8]

OR

Explain Collection Partitioning, source selection and query processing with respect to Distributed IR.                                                                                                                       [10]

What is parallel computing? Discuss performance measures of parallel computing.        [8]

How image analysis and image access accomplished in MULTOS Data Model.   [8]

Explain the Generic Multimedia Indexing Approach.                                 [8]

OR

Discuss the application of the GEMINI approach for Two- dimentional Color Images.   [8]

Discuss Uncertainty, Proximity, and Weights in Query Expressions. [8]

QII) a) Discuss different forms of searching the web. Explain with proper example.                                                                                                               [8]

b) Explain the crawler-indexer architecture.                                                         [8]

OR

Q12) a) Write short notes on:Web Data Mining.                                                              [8]