Collaborative QoS Prediction via Feedback-Based Trust Model

With the development of Service Oriented Computing and the explosive growth of services, QoS (Quality of Service) based service computing, e.g., QoS-based selection and composition, is becoming more and more important. A common premise of previous research is that the QoS values of services to target users are supposed to be all known, while many QoS values are unknown in reality. This paper presents a trust-aware Collaborative Filtering approach to predict such unknown values for QoS-based service computing. Compared with existing methods, the proposed approach has two new features: 1) prediction feedback is explored and utilized to improve the performance of QoS prediction, 2) the proposed trust model is complemental to the state-of-the-art Collaborative Filtering based prediction approaches. An extensive performance study based on a public dataset demonstrates the effectiveness of the proposed approach.