Abstract: Cluster analysis is one of the most important fields of data mining.K-means algorithm has important value in data mining.K-means must be given the number of clusters and it forms local convergence easily.So a new clustering algorithm，K-means based on the Shared Nearest Neighbor（KSNN），is designed.KSNN finds the core nodes of the data to get the number of clusters and takes it as the parameter for K-means automatically.It conquers the problem that the number of clusters to K-means must be defined by humans，meanwhile it has better global convergence.Experimental results show that KSNN is more effective than K-means，Particle Swarm Optimal（PSO） and multiseed core algorithm（MCA）.