K-means Multiple Clustering Research Based on Pseudo Parallel Genetic Algorithm

摘要

As K-means Clustering Algorithm is sensitive to the choice of the initial cluster centers and it is difficult to determine the cluster number and it is easy to be impacted by isolated points, propose the K-means multiple Clustering Method Based on Pseudo Parallel Genetic Algorithm. In the method, adopt the strategy of Variable-Length Chromosome real-coded. Through the introduction of chromosome retreading and focusing operator, K-means algorithm can be perfectly combined with pseudo-parallel genetic algorithm. For the dynamic and directed adjustment of migration rate with the evolutionary process, we have improved the migration rate of PPGA. The results of repeated experiment show that the method can effectively solve the previous problem and it is a practical and effective clustering algorithm.