+1 vote
in Data Structures & Algorithms I by (88.2k points)
How many hashes will be needed for calculating Jaccard index with an expected error less than or equal to 0.05?

(a) 100

(b) 200

(c) 300

(d) 400

My doubt stems from Hash Tables topic in chapter Hash Tables of Data Structures & Algorithms I

The question was posed to me by my college professor while I was bunking the class.

1 Answer

+1 vote
by (672k points)
selected by
 
Best answer
The correct answer is (d) 400

Easy explanation - The expected error for estimating the Jaccard index using MinHash scheme for k different hash functions is O (1/k½). 400 hashes will be needed for calculating Jaccard index with an expected error less than or equal to 0.05.

Related questions

Welcome to TalkJarvis QnA, a question-answer community website for the people by the people. On TalkJarvis QnA you can ask your doubts, curiosity, questions and whatever going in your mind either related to studies or others. Experts and people from different fields will answer.

Categories

...