摘要
分析了并行遗传算法的主从并行模型、粗粒度并行模型和细粒度并行模型等3种并行模型;重点对粗粒度并行遗传算法的异步迁移策略和同步迁移策略进行了比较研究,以K-中心聚类作为实验背景,说明不同的迁移策略对K-中心聚类的效果有一定影响,并得出了异步迁移策略的性能和聚类效果优于同步迁移策略的性能和聚类效果。
By analyzing three parallel models for parallel genetic algorithm: master-slave parallel models, coarse parallel model, and fine parallel model, asynchronous policy of migration and synchronous policy of migration of parallel genetic algorithm are stressed,K-Median Cluster as condition of experiment is to show Migration Strategy to have effect on it, at the same time ,proving asynchronous policy of migration is more effective than synchronous policy of migration in effect of cluster.
出处
《重庆工商大学学报（自然科学版）》
2005年第3期285-288,共4页
Journal of Chongqing Technology and Business University:Natural Science Edition
