J4 ›› 2010, Vol. 45 ›› Issue (7): 28-33.

• Articles • Previous Articles     Next Articles

An improved K-means algorithm by weighted distance based on maximum between-cluster variation

ZHANG Xue-feng1, LIU Peng1,2   

  1. 1. School of Information Management and Engineering, Shanghai University of Finance & Economics, Shanghai 200433, China;
     2. School of Continuing Education, Shanghai University of Finance & Economics, Shanghai 200080, China
  • Received:2010-04-02 Online:2010-07-16 Published:2010-09-06

Abstract:

To find natural clusters, the criterion function was improved by being defined as the weighted sum of the squared error. The way each point being assigned to the centroid in the iteration of the K-means algorithm was also modified: each point was assigned to the centroid that had minimum weighted distance. The weight was related with the number of points in each cluster, and the parameter of weighted distance was optimized by maximizing the between-cluster variation. Experimental results showed that the improved K-means algorithm significantly enhanced the clustering quality by reducing the probability of larger cluster’s being broken.
 

Key words: K-means algorithm; clustering; between-cluster variation; weighted distance

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!