Scribd
Upload
11 views

c3-paper(1)

Uploaded by

Manish Nagpure
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
11 views

c3-paper(1)

Uploaded by

Manish Nagpure
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

Subject Information Retrieval C3 Date: 26/11/2022 Time: 6.

45
pm to 8.15 pm MM: 25
Write down answer of all question in one or maximum two sentences in google form

1. Calculate Retrieval Status Value (RSV) for term_2 (based on given term-document
matrix) using Binary Independence Model for a query Q .(2.5)

Suppose for each term corresponding document value (1 or 0) showed that given term
is present or not. Relevant row (second row) show that Documents are relevant or not
for query Q

Total number of Docs = 8

term/D Doc_1 Doc_2 Doc_3 Doc_4 Doc_5 Doc_6 Doc_7 Doc_8


oc

Releva 0 1 0 1 0 1 0 1
nt

term_1 0 1 1 1 0 1 0 0

term_2 1 0 1 0 0 1 0 1

term_3 1 1 1 1 0 0 0 1

term_ 1 1 1 0 1 0 1 0
4

term_5 0 1 0 0 0 0 0 1

term_6 0 0 1 0 0 1 0 0

2. What is main idea behind BIM (Binary Independent Model). (1 marks)

3. For a given matrix A=[[2, 0, 1], [0, 1, 0], [0,0, 0] ] , identify the values of U, ∑ and V'.
(1+0.5+1 marks)

4. For a query, number of total relevant Documents are 17, number of total retrieved
relevant document are 10 and Total number of retrieved documents are 14, Calculate
Matthews Correlation Coefficient(mcc). (1.5 marks)

5. Calculate P(A/B) if P(B/A) = 0.2 P(A) = 0.3 and P(B) = 0.3 using Bayes theorem. (1 marks)

6. For given ` Terms-Document matrix` (Each column is corresponding to a Document


and each rows are correspond to frequency of terms present in corresponding
Documents) find out the pair of two different terms which has largest co-occurrence.
(1 marks)
Term/Document Docum_1 Docum_2 Docum_3

term_1 0 0 1

term_2 2 0 1

term_3 0 1 1

7. Apply edit distance (Levenshtein distance) b/n given two words. Write down edit
distance matrix of final state. (2 marks)

a. BEEGEGD b. FEABDFD.

8. What is the interpretation of diagonal of co-occurrence matrix. (0.5 marks)

9. Compute tf-idf vector for term_1 using steps 1: find term_frequency for documents
step 2 normalize terms_frequency for document to unit length. step 3 find idf using
formula log(n/n_i) step 4: multiply tf and idf. Term-Document graph is given below. (2.5
marks)

Term/Docu Docum_1 Docum_2 Docum_3 Docum_4 Docum_5


ment

term_1 3 14 5 1 11

term_2 6 7 2 16 17

term_3 10 17 14 15 3

term_4 1 11 0 9 9

10. What is use of n-gram overlapping in edit distance algorithm. (1 marks)

11. A fair coin is tossed, What is the a priori probability of landing a head? (0.5 marks)

12. What is different between contextual and global word embeddings. (1 marks)

13. Find Jaccard coefficient for given two sets.

set1 = { F, J, P, W, O, X, C, E, Y, Q, } set2 = { F, J, K, N, W, H, M, V, L, Q, }. (1 marks)

14. Calculate modified query vector using Rocchio Algorithm, based on given vectors. (2
marks)

original query vector =

0 0 1 0 0

Average of sum of known relevant Document-vectors =


0 1 1 0 1

Average of sum of known irrelevant Document-vectors=

2 1 0 2 2

Let alpha = 0.75, bita = 0.70, gama = 0.25

15. Write down all trigram index of word LXLELI . (0.5 marks)

16. Write down one challenge in computing Recall. (0.5 marks)

17. Let there are 6 sets A, B, C, D, E, F length of set A len(A) = 100, len(B)=110, len(C)=120,
len(D)=130, len(E)=140, len(F)=150 What will be order of execution for query Q= (A or F)
and (B or E) and (C or B). (1 marks)

18. Suppose, you have some set of books in the IIIT Allahabad library, let’s say 5 books
are there and among them 3 books are of IR and 2 book are of Computer Vision. Each
book has book id (BookID) and some set of words where they belongs to either IR or
Computer Vision, please see the below mentioned table for more detail.

Now, from this information compute the prior of P(B) and P(B'). (1.5 marks)

19. What is Thesaurus-based Query Expansion. Write down one example. (1 marks).

You might also like