Total No. of Questions : 6] SEAT No.

8
23
P273 [Total No. of Pages : 2

ic-
B.E/INSEM/APR-600

tat
5s
B.E. (Information Technology) (Semester - II)

1:0
414464B : INFORMATION STORAGE AND RETRIEVAL

02 91
3:4
(2015 Pattern) (Elective - III)
0
01
Time : 1 Hour] 4/0 13 [Max. Marks : 30
Instructions to the candidates:
0
3/2
.23 GP

1) Solve Q.1 or Q.2, Q.3 or Q.4, Q.5 or Q.6.

2) Figures to right indicate full marks.
E
80

8
C

23
ic-
Q1) a) You are developing a text processing system for use in an automatic
16

tat
retrieval System. What are different steps of conflation algorithm. [6]
8.2

5s
.24

1:0
91
b) Draw and explain IR system block diagram. [4]
49

3:4
30
01

OR
01
02
3/2
GP

Q2) a) Differentiate between data retrieval and information retrieval. [6]

4/0
CE
80

b) Explain Luhn’s idea in detail with diagram. [4]

8
23
.23

ic-
16

tat
8.2

5s

Q3) a) Explain single pass algorithm with example. [6]

.24

1:0
91
49

3:4

b) Write a short note on probabilistic model, vector model. [4]

30
01
01
02

OR
3/2
GP
4/0

Q4) a) Explain Single Link algorithm with example. [6]

CE
80
.23

b) Explain Latent Semantic Indexing with a suitable example. [4]

16
8.2
.24

P.T.O.
49
Q5) a) Define and explain following terms - Precision & Recall. [6]

8
23
b) Explain the terms Harmonic mean, E measure. [4]

ic-
tat
OR

5s
1:0
Q6) a) Explain User oriented measures for evaluating performance of IR system.

02 91
3:4
[6]

0
01
b) 4/0 13
Define and explain following terms : [4]
0
3/2
.23 GP

i) MRR.
E

ii) NDCG.
80

8
C

23
ic-
16

tat

8.2

5s
.24

1:0
91
49

3:4
30
01
01
02
3/2
GP
4/0
CE
80

8
23
.23

ic-
16

tat
8.2

5s
.24

1:0
91
49

3:4
30
01
01
02
3/2
GP
4/0
CE
80
.23
16
8.2
.24

B.E/INSEM/APR-600 2
49
Total No. of Questions : 6] SEAT No. :

8
23
P1423 [Total No. of Pages : 2

ic-
BE/Insem./APR-259

tat
1s
B.E. (Information Technology)

2:0
INFORMATION STORAGE AND RETRIEVAL

01 91
3:4
(2015 Course) (Semester -II) (Elective - III) (414464B)

0
91
3/0 13
0
Time : 1 Hour] [Max. Marks :30
3/2
.23 GP

Instructions to the candidates:

1) Solve Q.1 or Q.2, Q.3 Or Q.4, Q.5 or Q.6.
E
81

8
C

23
2) Figures to right indicate full marks.

ic-
16

tat
Q1) a) What is a document representative? Explain with a suitable example. [5]
8.2

1s
b) Draw and explain IR system block diagram. [5]
.24

2:0
91
49

OR
3:4
30
91

Q2) a) Explain steps in conflation algorithm using a suitable example. [5]

01
01

b) Write a short note on XML data structures used in IR systems. [5]

3/2
GP
3/0
CE
81

8
Q3) a) Find the similarity of the following query with documents - D1, D2, D3

23
.23

using vector model. [5]

ic-
16

tat
Query Keywords
8.2

1s

Q mouse, dog
.24

2:0
91

Document Text Terms

49

3:4
30

D1 mouse mouse bee mouse bee

91

D2 dog bee dog hog mouse bee dog hog

01
01
3/2

hog dog mouse dog

GP
3/0

D3 cat gnu dog eel fox Cat dog eel fox gnu
CE
81

b) Show how single link clusters may be derived from the dissimilarity
.23

coefficient by thresholding it. [5]

16
8.2

OR
.24

BE/Insem.-259 1 P.T.O.
49
Q4) a) Explain Latent Semantic Indexing with a suitable example. [5]

8
23
b) Compare Neural network-based retrieval and Fuzzy set retrieval methods.

ic-
[5]

tat
1s
2:0
01 91
3:4
Q5) a) Define and explain following terms - Precision & Recall. [5]

0
91
b) 3/0 13
Define and explain following terms - [5]
0
3/2
.23 GP

i) MRR
E

ii) NDCG
81

8
C

23
OR

ic-
16

tat
8.2

1s
Q6) a) Define and explain following concepts. [5]
.24

2:0
91
49

3:4
i) Cross fold validation.
30
91

ii) User oriented measures for performance evaluation.

01
01
3/2
GP

b) Define and explain following concepts related to visualization. [5]

3/0
CE

i) User relevance judgement.

81

8
23
.23

ic-
ii) Interface support for search process.
16

tat
8.2

1s

EEE
.24

2:0
91
49

3:4
30
91
01
01
3/2
GP
3/0
CE
81
.23
16
8.2
.24

BE/Insem.-259 2
49
Total No. of Questions : 10] SEAT No. :

8
23
P4009 [5561]-716
[Total No. of Pages : 2

ic-
B.E. (Information Technology)

tat
8s
INFORMATION STORAGE & RETRIEVAL

6:4
(2015 Course) (414464B) (Semester - II) (Elective - III)

01 91
3:4
0
Time : 2½ Hours] [Max. Marks : 70

91
4/0 13
Instructions to the candidates:
0
1) Neat diagrams must be drawn wherever necessary.
5/2
.23 GP

2) Figures to the right side indicate full marks.

3) Assume suitable data if necessary.
E
82

8
C

23
Q1) a) Differentiate between data retrieval and information retrieval. [6]

ic-
b) List with definition different measures of association. [4]
16

tat
OR
8.2

8s
Q2) a) Compare Boolean and vector model. [6]
.24

6:4
b) List and explain steps of conflation algorithm. [4]
91
49

3:4
30

Q3) a) Explain the terms Harmonic mean, E measure, R precision, Precision

91

histogram. [5]
01
01

b) Explain basic concept for Information Retrieval. Draw IR system block

5/2
GP

diagram. [5]
4/0

OR
CE
82

8
Q4) a) Dissimilarity matrix is given as follows. [5]

23
.23

1
ic-
16

2 0.6
tat
8.2

3 0.6 0.8
8s
.24

4 0.9 0.9 0.7

6:4
91

5 0.9 0.6 0.6 0.9

49

3:4

6 0.5 0.5 0.9 0.5 0.5

30
91

1 2 3 4 5 6
01
01

Threshold 0.4, 0.6, 0.8, 0.9.

5/2

Apply single link algorithm and calculate cluster for above 6 objects.
GP
4/0

b) Explain signature structure in detail. [5]

CE
82

Q5) a) What do you understand by multimedia query language? Explain various

.23

query predictors. [9]

16

b) Explain GEMINI approach for multimedia IR. [9]

8.2

OR
.24

P.T.O.
49
Q6) a) Describe multimedia data support in commercial DBMS. [9]

8
23
b) Describe the architecture of distributed IR. [9]

ic-
tat
8s
Q7) a) What is web crawling? Explain techniques used by web crawlers to crawl

6:4
the web. [8]

01 91
3:4
b) Write short note on web data mining. [8]

0
91
4/0 13 OR
0
5/2
Q8) a) Discuss the challenges involve in web search engine. [8]
.23 GP

b) What is page ranking? Calculate page rank of following web pages.

E
82

Assume damping factor 7.0. [8]

8
C

23
ic-
16

tat
8.2

8s
.24

Q9) a) Define Recommender system? Explain in brief collaborative filtering.[8]

6:4
91
49

3:4
b) Explain semantic web in detail. [8]
30
91

OR
01
01

Q10)a) Write a note on "Ontology languages for semantic web". [8]

5/2
GP

b) Explain the method for extracting data from text. [8]

4/0
CE
82

8
23

.23

ic-
16

tat
8.2

8s
.24

6:4
91
49

3:4
30
91
01
01
5/2
GP
4/0
CE
82
.23
16
8.2
.24

2
49

[5561]-716
Total No. of Questions : 10] SEAT No. :
P3260 [Total No. of Pages :2
[5461] - 298

1
8:2
B.E. (Information Technology)

3:3
INFORMATION STORAGE AND RETRIEVAL

81
1/2 1
9
(2012 Course) (End Sem.)(Semester - II) (414463 C) (Elective - III)

01
8 2 30
.23 01
Time : 2½ Hours] [Max. Marks :70
2/1
Instructions to the candidates:
8.2 G
P

1) All questions are compulsory.

.24 CE

2) Neat diagram must be drawn wherever necessary.

16

3) Figures to the right indicate full marks.

4) Assume suitable data, if necessary.

1
8:2
49

3:3
Q1) a) Write difference between Data retrieval and Information Retrieval. Define
81
91
Index term. [5]
01
30
1/2

b) Let [5]
01
2/1

Document 1 = {CPU, Keyboard, RAM, VGA, SMPS, USB, CD-ROM,

GP
82

Printer}
.23
CE

Document 2 = {CPU, VGA, Simulator, OS, Video, USB, Printer, Scanner,

16

Compiler}.
8.2

Find the similarity between two documents using 1

.24

8:2

i) Simple matching function

49

3:3

ii) Dice’s Coefficient

81
91
01

iii) Jaccard’s coefficient

30
1/2

iv) Cosine coefficient

01
2/1
GP

v) Overlap coefficient.
82
.23

OR
CE
16
8.2

Q2) a) Explain Riccho’s algorithm. [5]

.24

b) What is clustering? Explain the use of clustering in IR. [5]

49

[5461]-298 1 P.T.O.
Q3) a) Explain Vector model in detail. [5]
b) Explain ontology and ontology life cycle. [5]
OR

1
8:2
Q4) Discuss cluster based retrieval strategy. Also explain how to define cluster

3:3
representative. [10]

81
1/2 1
9
01
8 2 30
Q5) a) How queries are processed in distributed IR?
.23 01 [8]
b) Explain GEMINI approach for multimedia IR. [8]
2/1
P
OR
8.2 G

Q6) a) Write a short note on MULTOS. [8]

.24 CE
16

b) What are the issues in distributed IR computing? Explain techniques

used to address these issues. [8]

1
8:2
49

3:3
Q7) a) What are meta crawlers? Explain with example. [8]
81
91
b) Explain centralized crawler indexer and harvest distributed architecture
01
30

of search engine. [10]

1/2
01

OR
2/1
GP
82

Q8) a) What is meant by web crawling? Explain processing steps in web crawling.
[6]
.23
CE

b) Write a short note on web data mining. [6]

16
8.2

c) What is robo exclusion? Explain structure of robo.txt. [6]

1
.24

8:2
49

3:3
81
91

Q9) a) Define recommender system? Explain in brief collaborative filtering. [8]

01

b) Explain the method for extracting data from text. [8]

30
1/2

OR
01
2/1

Q10)a) Discuss trends and research issues involved in web. [8]

GP
82

b) Explain in details web usage mining. [8]

.23
CE
16
8.2


.24
49

[5461]-298 2
Total No. of Questions : 10] SEAT No. :

P3191 [Total No. of Pages : 2

4
[5354]-708

9:1
3:5
B.E. (Information Technology)

81
5/2 1
INFORMATION STORAGE AND RETRIEVAL

9
01
8 2 30
(Elective - III) (2012 Pattern) (End-Sem)
.23 01
Time : 2½ Hours] [Max. Marks :70
2/0
Instructions to the candidates:
8.2 G
P

1) All questions are compulsory.

.24 CE

2) Neat diagram must be drawn wherever necessary.

16

3) Figures to the right indicate full marks.

4) Assume suitable data, if necessary.

4
9:1
49

3:5
Q1) What is term weighting? Explain the TF-IDF scheme to calculate the weight
81
91
of index term. Find the weight of following terms. [10]
01
30
5/2

Document Text Terms

01
2/0

D1 SNMP,SNMP,FTP SNMP,FTP
GP
82

D2 HTTP,FTP,HTTP,ARP, SNMP,FTP,HTTP,ARP
.23
CE

HTTP,SNMP,HTTP
16
8.2

D3 NIC,INTERNET,HTTP, NIC,HTTP,PROTOCOL,
4
.24

PROTOCOL,HUB HUB,INTERNET
9:1
49

3:5

OR
81
91

Q2) a) Explain Single Link algorithm with example. [6]

01
30
5/2

b) Write a short note on probabilistic model. [4]

01
2/0
GP
82
.23

Q3) a) Explain inverted File structure with the help of diagram. State how it
CE
16

is useful in implementation of Information Retrieval System. [5]

8.2

b) Justify how vector model is used to retrieve partial matching documents.

.24

[5]
49

P.T.O.
OR

Q4) What is signature file? Describe false drop and search optimization using

4
9:1
signature files. Justify with example [10]

3:5
Q5) a) What is the need of distributed IR? Draw and explain architecture of

81
5/2 1
9
distributed IR system. [8]

01
8 2 30
b) What do you mean by collection partitioning & source selection in
Distributed IR? [8]
.23 01
2/0
P

OR
8.2 G
.24 CE

Q6) a) Explain with example some of the predicates used in multimedia query
16

language. [8]
b) What is indexing? How to index multimedia objects? [8]

4
9:1
49

3:5
Q7) a) Compare any two search engines with respect to features they support.
81 [6]
91
b) What is page ranking? Explain with example how to calculate a page
01
30

rank of web page? Comment on significance of dampening factor.[12]

5/2
01
2/0

OR
GP
82
.23

Q8) a) Explain importance of web crawling in searching. Describe with

CE

example two commonly used crawling strategies by web crawler.[10]

16

b) What are the challenges while searching the web? [8]

8.2

4
.24

9:1
Q9) a) How to collect and integrate specialized information on the web? [8]
49

3:5

b) How semantic web is useful in web searching? [8]

81
91

OR
01
30
5/2
01

Q10)a) Explain content based recommender system with example. [8]

2/0

b) What are the practical issues in collaborative filtering? [8]

GP
82
.23
CE
16
8.2
.24

vvvv
49

[5354]-708 2
Total No. of Questions :6] SEAT No. :

P 63 [Total No. of Pages : 2

APR-17/BE/Insem-73
B.E. (Information Technology) (Semester -II)
INFORMATION STORAGE AND RETRIEVAL
(2012 - Pattern)
Time : 1 Hour] [Max. Marks :30
Instructions to the candidates:
1) Answer Q1 or Q2, Q3 or Q4, Q5 or Q6
2) Neat diagram must be drawn wherever necessary.
3) Figures to the right indicate full marks.
4) Assume suitable data if necessary

Q1) a) Explain working of conflation algorithm in detail. Justify use of this

algorithm in information retrieval. [6]

b) Write a short note on matching coefficients. [4]

OR

Q2) a) Why single pass algorithm is better than Rocchio's Algorithm? [10]
Form the document cluster of following document term matrix using
single pass clustering algorithm.
Consider
Membership function : Sum of product
centroid calculation function : Average
Threshold = 11
D1 D2 D3 D4 D5
T1 1 1 0 1 1
T2 2 1 2 3 0
T3 3 0 1 0 1
T4 2 2 0 3 0
T5 2 2 1 2 1

P.T.O
Q3) a) Compare boolean model and vector model. Explain how vector model
can be used to retrieve partial matching documents. [6]

b) What are inverted files? Explain how these file can be used to answer
Boolean queries. [4]

OR

Q4) a) Explain working of suffix tree. Construct suffix tree for following example
"This is a text. A text has many words. Words are made from letters."[6]

b) Explain working of signature files with example? [4]

Q5) a) Wirte a note on user oriented measures used for evaluating the performance
of any retrieval system. Also explain their significance. [6]

b) Explain the term ontology creation [4]

OR

Q6) a) Write a note on " Ontology languages for semantic web". [5]

b) Explain the trade-off between precision and recall. [5]

iii

BE/Insem-73 2
Total No. of Questions : 10] SEAT No :

6
R3
P3142 [5154]-708
[Total No. of Pages :2

VE
B.E.(I.T.)

ER
INFORMATION STORAGE AND RETRIEVAL

S
:02
(2012 Course) (Elective-III) (414463 C) (Semester-II)

17 91
:38
/20 30
13
Time : 2½ Hours 1 Max.Marks:70
Instructions to the candidates:
22 P0

1) All questions are compulsory.

2) Neat diagrams must be drawn wherever necessary.
9.2 EG

/05

3) Figures to the right indicate full marks.

36
4) Assume suitable data, if necessary.
C

ER
34

RV
Q1) a) Show how single link clusters may be derived from the dissimilarity
coefficient by thresholding it. [5]

SE
1.3

b) You are developing a text processing system for use in an automatic

.14

:02
91
retrieval System. Explain the following parts: [5]
14

:38
Removal of high frequency words.
30
13

Suffix stripping.
01

Detecting equivalent stems.

17

OR
/20
GP
/05

Q2) a) Find the similarity of following query with D1,D2,D3, using vector model.
CE

36
22

[6]

ER
34

Query keywords
RV
9.2

SE
1.3

q ant, dog
.14

:02

document Text Terms

91
14

:38

D1 ant ant bee ant bee

30
13

D2 dog bee dog hog dog ant dog ant bee dog hog
01
17
/20

D3 cat gnu dog eel fox cat dog eel fox gnu
GP
/05

b) Write a short note on user oriented measures to evaluate the performance

CE
22

of the system. [4]

34
9.2

Q3) a) Writr a note on “Ontology based information sharing”. [5]

b) Explain the concept of hash addressing. [5]
1.3

OR
.14

[5154]-708 1 P.T.O.
14
Q4) Consider a reference collection and its set of example information request. If

6
R3
q is the information request and a set Rq= (d3,d5,d9,d25,d39,d44,d50,d70,

VE
d80,d120). Now consider new retrieval algorithm has been designed and has

ER
been evaluated for information request q returns, ranking of the documents
in the answer set as. [10]

S
1.d120 6.d9 11.d38

:02
17 91
2.d84 7.d58 12.d48

:38
/20 30
3.d50 8.d129 13.d230

13
4.d6 9.d143 14.d113
1
5.d8 10.d25 15.d3
22 P0

The documents that are relevant to the query q are underlined. Calculate
9.2 EG

/05

precision and recall for the documents that are relevant to the query q.

36
Q5) a) Describe the architecture of distributed IR. [8]
C

ER
34

b) What do you understand by multimedia query language? Explain various

RV
query predicates. [8]
OR

SE
1.3

Q6) a) What are the issues in distributed IR computing? Write the techniques
.14

:02
91
used to address these issues. [8]
14

:38
b) Write a note on MULTOS. [8]
30
13

Q7) a) Write a short note on web data mining. [8]

01
17

b) What is page ranking? Calculate page rank of following web pages.

/20
GP

Assume damping factor 7.0. [10]

/05
CE

36
22

ER
34

RV
9.2

OR
SE
1.3

Q8) a) Explain centralized and distributed architecture of a search engine. [10]

.14

:02

b) What is web crawling? Explain the techniques used by web crawlers to

91

crawl the web. [8]

14

:38
30

Q9) a) What is content based recommendation? [8]

13

b) Explain semantic web in details. [8]

01
17
/20
GP

OR
/05

Q10)a) Define Recommender system? Explain in brief collaborative filtering [8]

CE

b) Discuss trends and research issues involved in web. [8]

22
34

S S S
9.2
1.3
.14

[5154]-708 2
14
Total No. of Questions : 10] SEAT No. :
P3647 [4959]-1138
[Total No. of Pages :2

B.E.(Information Technology)
c: INFORMATION STORAGE AND RETRIEVAL
(2012 Course) (Semester-II)(Elective-III) (414463)
Time :2½Hours] [Max. Marks : 70
Instructions to the candidates:
1) All questions are compulsory.
2) Neat diagrams must be drawn wherever necessary.
3) Figures to the right indicate full marks.
4) Assume suitable data if necessary.

Q1) a) Explain Luhn’s idea in details. [5]

b) What is clustering? Explain the use of clustering in IR. [5]
OR
Q2) Clusters the documents using single pass clustering algorithm for the following
example. Threshold value is 10 [10]
Terms in document
T1 T2 T3 T4 T5
Doc 1 1 2 0 0 1
Doc 2 3 1 2 3 0
Doc 3 3 0 0 0 1
Doc 4 2 1 0 3 0
Doc 5 2 2 1 5 1

Q3) a) Explain Boolean model in detail [5]

b) Give the difference between suffix array and suffix tree. [5]
OR

[4959]-1138 1 P.T.O.
Q4) a) Explain the term precision and recall and calculate the same for the
following example [5]
A set of relevant documents for query
q = {d3,d7,d8,d11,d14,d19,d23,d25}
A new retrieval algorithm returns following answer set
= {d1,d2,d3,d7,d9,d10,d14,d20,d23,d24,d25}
b) Explain the terms Harmonic mean, E measure, R precision, Precision
histogram [5]

Q5) a) Explain GEMINI approach for multimedia IR. [9]

b) How are queries processed in distributed IR. [9]
OR
Q6) a) Write a note on MULTOS [9]
b) Describe multimedia data support in commercial DBMS. [9]
Q7) a) Discuss challenges involved in web searching. [8]
b) Explain crawler indexer architecture in details. [8]

OR

Q8) a) Write a note on characterizing the web [8]

b) What are meta crawlers? Explain with suitable example [8]

Q9) a) What is collaborative filtering? Discuss its advantages and disadvantages

[8]

b) Explain semantic web in details [8]

OR

Q10)a) Explain the method for extracting data from text [8]
b) Explain Collecting and Integrating Specialized Information on the web.
[8]

tu tu tu

[4959]-1138 2
Total No. of Questions : 10] SEAT No. :

P2072 [Total No. of Pages : 3

[5059] - 678
B.E. (Information Technology)
INFORMATION STORAGE AND RETRIEVAL
(2012 Pattern) (Elective - III)
Time : 2½ Hours] [Max. Marks :70
Instructions to the candidates:
1) All questions are compulsory.
2) Neat diagrams must be drawn wherever necessary.
3) Figures to the right indicate full marks.
4) Assume Suitable data, if necessary.

Q1) Why single pass algorithm is better than Rocchio’s Algorithm? Form the
document cluster of following document term matrix using single pass
clustering algorithm. Consider [10]
Membership Function: Sum of product
Centroid calculation Function: Average
Threshold = 11
Dl D2 D3 D4 D5
T1 1 1 0 1 1
T2 2 1 2 3 0
T3 3 0 1 0 1
T4 2 2 0 3 0
T5 2 2 1 2 1

OR

Q2) a) Explain working of suffix tree. Construct suffix tree for following
example. [6]
“This is a text. A text has many words. Words are made from letters.”
b) Write a short note on matching coefficients. [4]
P.T.O.
Q3) a) Write a note on “Ontology languages for semantic web”. [5]
b) Write a note on “cluster based retrieval”. [5]

OR

Q4) Consider a reference collection and its set of example information request.
If q is the information request and a set [10]
Rq=(d3, d5, d9, d25, d39, d44, d50, d70, d80, dl20). Now consider new
retrieval algorithm has been designed and has been evaluated for information
request q returns, ranking of the documents in the answer set as.
1. d120 6. d9 11. d38
2. d84 7. d58 12. d48
3. d50 8. d129 13. d230
4. d6 9. d143 14. d113
5. d8 10. d25 15. d3
The documents that are relevant to the query q are underlined. Calculate
precision and recall for the documents that are relevant to the query q

Q5) a) Describe the query processing in Distributed IR. [8]

b) Explain Data models used for Multimedia IR. [8]

OR

Q6) a) What do you mean by collection partitioning & source selection in

Distributed IR? [8]
b) Explain the generic multimedia Indexing Approach. [8]

Q7) a) What is Metasearcher? [6]

b) What is importance of page ranking? Calculate page rank of following
web pages. [12]

OR

[5059]-678 2
Q8) a) What are the challenges while searching the web? [12]
b) What is the role of crawler in web searching? Explain the strategies
used by web crawler. [6]

Q9) a) Define Recommender System? Explain in brief collaborative filtering.

[8]
b) Explain the method of extracting data from text. [8]

OR

Q10)a) Explain the concept of semantic web .How it is useful in web searching?
[8]
b) Explain in detail content based recommendation of documents. [8]

vvvv

[5059]-678 3