Reputation-enhanced similarity-based dynamic service discovery framework for service-based applications

Service-based applications are applications that make use of one or more autonomous computer-based entities owned by third parties known as services, to perform their functions. The identification of similar, reputed services during the execution of service-based applications to replace services in them that are no longer available is an important issue. This paper proposes a framework to support dynamic service discovery. The users subscribe with the framework and submit their request for services. The framework is integrated with a Web service repository, where the service providers publish their services with QoS specification. The framework also collects the user ratings of services. To discover services, first the framework finds the similarity between functional and quality characteristics of services in the repository and the user requests. Then it calculates the reputation score of the services based on the ratings collected. Based on the similarity score and reputation score, the framework lists and ranks a set of candidate services for replacement. The framework can perform service discovery both proactively and reactively. Thus the framework provides an effective, user-friendly and flexible approach for run time service identification for service-based applications.