基于免疫遗传算法的Web服务组合研究

Research on Immune Genetic Algorithm Based Web Services Composition

单个Web服务难以满足实际应用的需求,如何组合已有的服务,形成新的服务,已成为此领域的研究热点。本文将每一个可能的Web服务方案表示成一个Web服务组合图,将Web服务问题转化为Web服务组合图优化选取问题,进而提出了一种基于免疫遗传算法的、实现Web服务组合图优化选取的服务组合方法。该方法将Web服务组合图的最长路径作为抗原,每一个Web服务组合图对应一个抗体。实验结果证明了该方法的服务组合时间短,且服务组合效率高。

Single Web service just provides limited functionality,and can t meet the needs in practice. How to compose existing services to form new services has become an important research in Web service domain. Each and every possible Web service scheme is expressed as a Web services composition graph,thus converting the Web services problem into a graphically optimal selection problem. Then,to find the optimal solution quickly and accurately,a Web services composition method based on immune genetic algorithm is proposed to implement the optimal selection. This algorithm takes the longest path in the Web services composition graph as an antigen and every Web services composition graph as a corresponding antibody. Experimental results indicate this method not only shortens the needed time of composition, but also increases the efficiency of services composition.