Learning-based trust model for optimized web services selection

Abstract

With the proliferation of Web services, Quality of Services(OoS) becomes a essential factor to differentiate Web services. The current mechanism for specifying QoS in Web services still in its infancy and lacking with many important features of dynamic nature of Web services. Web services itself and QoS of Web services are dynamic enough which makes it hard to predict QoS of provider Web services. On the other hand, with the deployment of Web services increases in complex business application integration, it is getting inevitable that several provider Web services may have the same or similar functionalities each holds different Quality of Services(QoS). Due to the increasing number of Web services with the same or similar functionalities, it is getting more and more difficult for Web services consumers to select most suitable services for their businesses. In this paper, we suggested a mechanism for specifying QoS. The mechanism which we devised is based on a model named Trust Model. The Trust-Model introduced in this paper has mainly two functions: A mechanism for specifying QoS of Web services, a method for optimized selection of provider Web services among several similar providers but with different QoS. Our Trust Model is a function of historically gathered and learned QoS values, references or feedbacks from other services and honesty degree of this service. Client Web service satisfaction degree function was devised for evaluating our Trust Model. The proposed mechanism for specifying QoS is validated with feasible result by comparing conventional way and our suggested model.