Abstract: Finding the appropriate number of clusters in the absence of prior information is a hard and sensitive problem in clustering and data analysis. In this paper, we present a new cluster ...