A User-Centric Trust and Reputation Method for Service Selection

Our work focuses on selecting a trusted service from service consumer\´s perspective. First, we distinguish the two kinds of trust context to separate the ability of service as both a service provider and a service consumer in a service web; and we classify the reputation into two different categories. Second we generate a reputation metric model by improving Subject Logic. The trust decision model based on the metric model takes the reliability of direct experiences as well as indirect experiences into account. Third, by building a "bidirectional" feedback mechanism and pattern analysis of service peers\´ behaviors, we address the malicious feedback and dishonest offering faced by using the reputation based trust systems in the service-oriented environment. Finally, we demonstrate the effectiveness of our method on service selection.