Mining Clusters in Data Sets of Data Mining: An Effective Algorithm

Abstract

This paper propose a new clustering algorithm (GACR) based on genetic algorithm. The searching capability of genetic algorithms is exploited in order to search for appropriate cluster centers in the feature space such that a similarity metric of the resulting clusters is optimized. The chromosomes, which are represented as strings of real numbers, encode the centers of a fixed number of clusters. A chromosome reorganization method is proposed, which may effectively remove the degeneracy for purpose of more efficient search. A new crossover operator that exploits a measure of similarity between chromosomes is also presented. Adaptive probabilities of crossover and mutation are employed to prevent the convergence of the GA to a local optimum. The features of this algorithm are investigated and the performance is evaluated experimentally using real and synthetic datasets with K-means and GCA (10).The experimental result demonstrates that the GACR clustering algorithm has high performance, effectiveness and flexibility.