News for hackers. Stuff that matters.

A comment on Hashing for large-scale similarity.

Re:Hashing for large-scale similarity by carlmr on Monday February 11, 2019 @03:45AM (#19132922)
The Jaccard and Cosine similarity metrics are easy to grasp. I think the hashes could take some work.

Also you're counting up to k hash functions, but the element before the three dots is already element k. I think it should be 1,2,...,k, instead of 1,k,...,k.

[ Parent | on HN ]
Re:Hashing for large-scale similarity by mesuvash on Sunday February 17, 2019 @04:28PM (#19187033)
>>I think the hashes could take some work. Any suggestions or thing that are not clear?

Thanks for your feedback. I shall update the post accordingly.

[ Parent | on HN ]