0

I need to do K means clustering with the difference that i need maximize the difference between clusters.I can't find a package to do it.Writing the package myself is difficult.

Thank You

jogall
  • 651
  • 6
  • 21
user1932582
  • 61
  • 4
  • Strange, I thought the aim of K-means is to find clusters minimizing the variance intra-cluster and maximizing the variance inter-cluster. –  Jul 10 '14 at 08:44
  • Actually I am implementing a research paper which shows for classification purposes its better to maximize scatter between class matrix http://www.google.co.in/url?sa=t&rct=j&q=&esrc=s&source=web&cd=2&ved=0CCQQFjAB&url=http%3A%2F%2Fwww.researchgate.net%2Fpublication%2F3970380_Boosting_face_recognition_on_a_large-scale_database%2Ffile%2Fd912f50ca0ccc29667.pdf&ei=K1--U9etH86IuAT274LADQ&usg=AFQjCNGqckPwBoGckAXDNBJpdHlPTVp5RA&sig2=sRpKcAyvURczSeSN9j2K2Q&bvm=bv.70138588,d.c2E&cad=rja – user1932582 Jul 10 '14 at 09:40
  • Then it is, by definition, not k-means anymore. – Has QUIT--Anony-Mousse Jul 10 '14 at 17:09

0 Answers0