A modified version of the K-means algorithm with a distance based on cluster symmetry

Mu-Chun Su, Chien-Hsing Chou
2001 IEEE Transactions on Pattern Analysis and Machine Intelligence  
AbstractÐIn this paper, we propose a modified version of the K-means algorithm to cluster data. The proposed algorithm adopts a novel nonmetric distance measure based on the idea of ªpoint symmetry.º This kind of ªpoint symmetry distanceº can be applied in data clustering and human face detection. Several data sets are used to illustrate its effectiveness.
doi:10.1109/34.927466 fatcat:yqwavmc4y5eyxeitc77oaudqaa