Abstract:

Traditional K-eans algorithm of the initial clustering center is randomly generated, which can lead to roduce very big volatility clustering results. In order to solve this problem, We propose a algorithm named lustering algorithm based on min-istance Product. With the method of sampling, CAMDP(Clustering Algorithm ased on Min-istance Product) produces selected point which has minimum product of distances between itself nd all other initialized clustering centers, which improves the selecting of the initial value of the K-eans lgorithm, avoiding the random selected clustering centers. The results show that the topological feature is onsidered and the attributes of vertex are taken into account, which let the improved K-eans provide the strong upport to the division of community.

Key words: community structure, clustering, social relations, clustering centers