CODEWITHSUNDEEP
cluster-analysisclustering-key
Prasad
Prasad

Reputation: 69

Which clustering algorithm is suitable for clustering geographical locations?

I am developing an application which works similar to Tinder. I am guessing Tinder groups closest results by running clustering algorithm. In my application have to similarly group data based on geographical location. I might have to run clustering based on many inputs, so it has to be efficient. Please suggest suitable algorithm for it.

Upvotes: 0

Views: 139

Answers (2)

Prasad
Prasad

Reputation: 69

For somebody who is looking for similar solution, There is a good answer about fast similarity search algoritms on quora https://www.quora.com/What-are-some-fast-similarity-search-algorithms-and-data-structures-for-high-dimensional-vectors/answer/Raghavendran-Balu?srid=hYuT

I found R-tree to be most suited for my application. There is a good github project for R-tree https://github.com/davidmoten/rtree

Upvotes: 0

Has QUIT--Anony-Mousse
Has QUIT--Anony-Mousse

Reputation: 77454

There is no reason to group or cluster for a Tinder-like use case:

  • it's too expensive
  • it's too static
  • it does not add value (you can't just present a cluster to the user)

What you want to use is similarity search. Find other users that are a) nearby, b) recently online, c) have some interests in common, d) have not been recently shown.

Upvotes: 0

Related Questions