Automatically determining the number of Affinity Propagation Clustering using Particle Swarm
Xianhui Wang, Xuanping Zhang, Chunyi Zhuang, Zehao Chen, Zheng Qin · 2010
Affinity propagation (AP) is a new powerful clustering algorithm based message passing between data points. One of the major problems in clustering is the determination of the optimal number of clusters. In this paper, we propose a new approach called Affinity Propagation Clustering based Particle Swarm Optimization (PSO-AP), which using Particle Swarm Optimization (PSO) algorithm to determination of the optimal clustering number of AP. Our PSO-AP method is absolutely “automatic”. PSO-AP represents the issue of finding the optimal clustering number of AP as an optimization problem of searching optimal solution of the input “preferences” space. It evaluates the particles' fitness using clustering validation indexes. Bounded constraint strategy of PSO used supervisor-student model. Several experiments data sets are presented to illustrate the simplicity and effectiveness of PSO-AP.