0
$\begingroup$

I wrote a program to check the similarity of 40 documents. Firstly i use exact Jaccard method and secondly Minhash Jaccard method. I consider that the 2 methods differ if their values are differ more than 0,1. I also run the program for permutations 10,40,100,200,450. Which Jaccard limit must i choose to consider 2 documents as similar?

$\endgroup$
2
  • 1
    $\begingroup$ It's up to you – there's no standard value. $\endgroup$ – Yuval Filmus Jan 6 '17 at 18:34
  • $\begingroup$ ok thank you. And something else, if i choose bigger numbers for permutation will i get more accurate results about the documents similarity? $\endgroup$ – Lee Yaan Jan 6 '17 at 22:15

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.