If word count for every term doubles in one document:
5. Question 5 If word count for every term doubles in one document: 1 point If not using any smoothing, query likelihood would change for some queries. p(w…
-
-
Assume the same scenario as in Question 3, but using linear interpolation (Jelinek-Mercer) smoothing with \lambda = 0.5λ=0.5. Furthermore, you are given the following probabilities of some of the words in the collection language model:
4. Question 4 Assume the same scenario as in Question 3, but using linear interpolation (Jelinek-Mercer) smoothing with \lambda = 0.5λ=0.5. Furthermore, you are given the following probabilities of some…
-
Assume the same scenario as in Question 1. What is the average precision of both systems?
Question 3 Assume the same scenario as in Question 1. What is the average precision of both systems? 1 point AP(A) = 1/20 AP(B) = 1/5 AP(A) =…
-
Assume the same scenario as in Question 1. What is the recall of both systems?
Question 2 Assume the same scenario as in Question 1. What is the recall of both systems? 1 point R(A) = 1/40 R(B)= 2/40 R(A) = 9/10 R(B)=…
-
Suppose a query has a total of 4 relevant documents in the collection. System A and System B have each retrieved 10 documents, and the relevance status of the ranked lists is shown below: System A: [- + – – – – – – – -] System B: [+ + – – – – – – – -]
1. Question 1 Suppose a query has a total of 4 relevant documents in the collection. System A and System B have each retrieved 10 documents, and the relevance status…
-
What can you say about the precision-recall (PR) curve?
7. Question 7 What can you say about the precision-recall (PR) curve? 1 point It is always monotonically increasing. The ideal system should have the PR curve as…
-
For any ranked list of search results, precision at 10 documents is always higher than precision at 20 documents.
6. Question 6 For any ranked list of search results, precision at 10 documents is always higher than precision at 20 documents. 1 point False True
-
Can a retrieval system have an F1 score of 0.75 and a precision of 0.5?
5. Question 5 Can a retrieval system have an F1 score of 0.75 and a precision of 0.5? 1 point No Yes
-
Which of following is wrong about nDCG@k?
10. Question 10 Which of following is wrong about nDCG@k? 1 point It can be used to compare across queries. It discounts only top ranked documents. It…
-
Which of the following is NOT true about Cranfield evaluation methodology?
9. Question 9 Which of the following is NOT true about Cranfield evaluation methodology? 1 point It simulates real document collections. It simulates user queries. It does…