Study on a density peak based clustering algorithm

Abstract

The density peak based clustering algorithm is a recently proposed clustering approach. It uses the local density of each data and the distance to the nearest neighbor with higher density to isolate and identify the cluster centers. After the cluster centers are identified, the other data are assigned labels equaling to those of their nearest neighbors with higher density. This algorithm is simple and efficient. On condition that the cluster centers are identified correctly, it can generate very good clustering results. However, the results of this algorithm depend on a parameter in the local density calculation. In this paper we investigate the influence of the parameter on the clustering results through extensive experiments on several datasets. Our work can be useful in applying the density peak based clustering algorithm to practical tasks.