Genetic Clustering Algorithm Based on Dynamic Granularity

Abstract

From the view of granularity, this paper presents a genetic clustering algorithm based on dynamic granularity. In view of a parallel, random search, global optimization and diversity characteristics of genetic algorithm, it is combined with dynamic granularity model. In the process of granularity changing, appropriate granulation can be made by coarsening and refining the granularity, which can ensure clustering efficiency and quality of the algorithm. Experimental data show that the method effectively improves the clustering algorithm based on genetic algorithm local search ability and convergence speed.