2
$\begingroup$

I have a distance matrix that I want to cluster. What methods can I use to cluster the elements just based on the distance matrix (beside hierarchical clustering)?

Improve this question
$\endgroup$
3
  • $\begingroup$ I recommend you post clarifications as suggestions become available. What will work for you probably depends a great deal on your use case, but I understand that you might not know how to narrow the field before seeing some options. $\endgroup$
    – Stumpy Joe Pete
    Commented Jul 20, 2013 at 7:28
  • $\begingroup$ I would like to point you to the fact that you can always obtain a cases X dimensions data from a distance matrix by multidimensional scaling. $\endgroup$
    – ttnphns
    Commented Jul 20, 2013 at 7:33
  • 2
    $\begingroup$ similar question stats.stackexchange.com/questions/2717/… I vote to delete my answer since I have found similar question that already has an answer. $\endgroup$
    – sitems
    Commented Jul 20, 2013 at 8:48

1 Answer 1

Reset to default
0
$\begingroup$

There are plenty of "distance based" clustering algorithms that can be used for any kind of similarity function (metrical or not).

For example DBSCAN and OPTICS, but there are literally hundreds more.

Improve this answer
$\endgroup$

Not the answer you're looking for? Browse other questions tagged or ask your own question.