Abstract: K-means algorithm is sensitive to initial value,easy to fall into local minimum value.In response to these shortcomings,the idea of genetic algorithm is proposed based on genetic algorithm and k-means algorithm for hybrid clustering method.In order to test the performance of clustering algorithm,a set of experiments are conducted by using k-means algorithm and the improved algorithm,and the clustering results by the two algorithms are compared.It is showed that the clustering algorithm can effectively solve the clustering problem.

Key words: data mining, cluster analysis, genetic algorithm, k-means algorithm