Abstract: Kernel clustering is an effective algorithm which can deal with samples that have weak differences.On the basis that of new improved attribute importance under the theoryof rough set is applied to the kernel clustering algorithm.Before the samplesare optimized by the kernel function,their properties is processed by the reduction algorithmwhich is based on the attribute importance.At the same time,Information Entropyis introduced to improve the reduction algorithm.So the redundant attributes aredeleted and the optimum set of attributes is obtained;Then,the samples areclustered by K-means clustering algorithms,and the samples are divided intotheupper and lower approximate subsets of the corresponding cluster centers.Due tothe samples in approximate subsets having different influence on cluster,different weighs are designed to determine the new clustering centers.This paper adopts UCI data sets to test the performance ofthe algorithm.Through the comparison with traditional kernel clustering algorithmis shows that the proposed clustering algorithm improves the cluster result'saccuracy,reduces the complexity and shortens the convergence time significantly.

Key words: rough set, attribute reduction, attribute importance, information entropy, kernel clustering