A Max-Min clustering method for $k$-means algorithm of data clustering

Abstract

As it is known that the performance of the $k$-means algorithm fordata clustering largely depends on the choice of the Max-Mincenters, and the algorithm generally uses random procedures to getthem. In order to improve the efficiency of the $k$-means algorithm,a good selection method of clustering starting centers is proposedin this paper. The proposed algorithm determines a Max-Min scale foreach cluster of patterns, and calculate Max-Min clustering centersaccording to the norm of the points. Experiments results show thatthe proposed algorithm provides good performance of clustering.