| Popis: |
Selecting high quality services from available ones plays an important role in services composition. Trust and reputation mechanisms offer a promising way to solve the services selection problem. Current researches on this topic focus on gathering, storing and aggregating feedbacks which can be execution data by monitoring services or consumer ratings. However, besides the feedback itself, the reliability of feedbacks should be considered equally crucial. In this paper, different from previous work, we draw our attention to find trusted feedback submitters who are more likely to give reliable feedbacks. We proposed a service evaluation framework where a "Web of Trust" is built and maintained. Moreover, we presented a Kalman Filter based algorithm to reveal the trusted feedback submitters. The presented work in this paper aims to supplement the service evaluation and ultimately facilitate the services selection process. We validated our approach by both simulated and real-world evaluations. |
