基于混合协同进化算法的Web服务组合演化策略

为了在服务组合演化过程中高效地选择满足服务请求的Web服务，提出了一种基于混合协同进化算法的Web服务组合演化策略。首先，利用改进模糊C均值聚类算法将Web服务演化单元按应用分类；然后，利用带权值的粒子群算法对划分好的子群进行内部择优；最后，对各个子群进行协同进化，使得针对用户服务请求，能够选出最优Web服务演化组合。实验结果表明，混合协同进化算法无论在算法稳定性或是算法运算时间上都优于传统协同进化算法，且对于Web环境下大量的服务请求能够提供优质、高效的服务。

Web service evolution framework based on hybrid co-evolution algorithm

In order to select the Web services that meet the requests in the evolution of Web services efficiently,this paper proposed a Web services evolution strategy based on hybrid co-evolutionary algorithm.First of all,by means of the improved Fuzzy C-Means(FCM) clustering algorithm,this paper classified the Web services units into groups according to the application,and then made the internal prioritizing for each subgroup using Particle Swarm Optimization(PSO) with weights.Finally the co-evolution operation would be executed among the particle subgroups so as to fulfill the users’ requests and select the best Web services evolution combination.The experimental results show that the hybrid co-evolutionary algorithm is superior to the traditional co-evolutionary algorithms not only in stability,but also in operation time,and can provide good and efficient services to the service requests in the Web environment.