A K-medoids Clustering Algorithm with Initial Centers Optimized by a P System

In this paper an improved K-medoids algorithm by a specific P system is proposed which extends the application of membrane computing. The traditional K-medoids clustering results vary accordingly to the initial centers which are selected randomly. In order to conquer the defect, we improve the algorithm by selecting the k initial centers based on the density parameter of data points. P system is adequate to solve clustering problem for its high parallelism and lower computational time complexity. A specific P system with the aim of realizing the improved K-medoids algorithm to form clusters is constructed. By computation of the designed system, it obtains one possible clustering result in a non-deterministic and maximal parallel way. Through example verification, it can improve the quality of clustering.