Mining the Web of Trust for Web Services Selection

Trust and reputation mechanisms offer a promising way to solve the web services selection problem. Currently, many researches on this topic focus on collecting, storing and aggregating feedbacks which can be execution data from monitoring web services or user ratings reported by service consumers for web services evaluation. However, besides considering the evaluation methods, the reliability of feedbacks becomes equally crucial. It is pointed out that the accuracy of a web service's reputation evaluation will be reduced as unreliable feedbacks are used. Therefore, the elicitation of trusted feedbacks and feedbackers becomes very important. In this paper, we managed to elicit trusted users in our service registry through building and maintaining a "Web of Trust". We mined this web of trust by our proposed algorithm which is based on the Kalman Filter Algorithm and analyzed the trustworthiness for each user. Feedbacks reported by trusted users then should be considered more reliable than others to ultimately facilitate the web services selection process.