An Improved Fuzzy Guided Genetic Algorithm for the Selection of Web Services

Web services are integrated web-based applications that function over a protocol backbone. Web service selection is one of the most important steps in many of the web service applications such as UDDI registries and web service repositories. Each web service has its own functionality which matches the service request. QoS plays a major role in selecting web services in terms of qualitative measure. The problem addressed in this paper is to select the exact web service in a dynamic environment which matches the consumer request with optimum solution quality. At first, a three-parent genetic algorithm has been introduced in this paper and compared with the existing two parent GA. Next An Improved FGA is proposed that considers the user’s preference given for the quality attributes and looks for the services in dynamic composite services. As in FGA, the fuzzy controller adjusts the crossover and mutation rates accordingly and the Gaussian membership function is used in fuzzy engine to obtain finer granularity in fuzzy sets. As a result an exact service that matches the consumer’s quality requirements is obtained. Results were compared with the existing FGA and presented. [email protected] KeywordsComposite Services, Fuzzy controller, Gaussian Membership Function, QoS.