2

I have a pretty big 5868 x 5868 matrix of users vs users that have values that indicate similar behavior. So for example the [i,j] cell of user X and user Y would be some value like 10 if they did the same things 10 times, and 1 if they did the same things 1 time.

I tried running the cophenetic distances algorithm on SO and this is what I get, and I'm at a loss to see how one can go about picking apart this data more effectively.

Any thoughts or ideas would be greatly appreciated!

Cophonetic distances

Community
  • 1
  • 1
Rio
  • 14,182
  • 21
  • 67
  • 107
  • This one should be pushed over to http://www.crossvalidated.com as the question is more about interpretation of results rather than specifically a programming question. But there are litterally hundreds of clustering methods available through R. See the task view: http://cran.r-project.org/web/views/Cluster.html – Brandon Bertelsen Feb 28 '12 at 03:45

0 Answers0