A two-stage genetic algorithm for automatic clustering
Neurocomputing
Hi-index | 0.00 |
As K-means Clustering Algorithm is sensitive to the choice of the initial cluster centers and it's difficult to determine the cluster number, we proposed a K-means Clustering Method Based on Parallel Genetic Algorithm. In the method, we adopted a new strategy of Variable-Length Chromosome Encoding and randomly chose initial clustering centers to form chromosomes among samples. Combining the efficiency of K-means Algorithm with the global optimization ability of Parallel Genetic Algorithm, the local optimal solution was avoided and the optimum number and optimum result of cluster were obtained by means of heredity, mutation in the community, and parallel evolution, intermarriage among communities. Experiments indicated that this algorithm was efficient and accurate.