Reputation-based Trust Framework for Service Oriented Environments

Abstract

We investigate the problem of establishing trust in service-oriented environments. We focus on providing a reputation framework that would enable trust-based interactions with and amongst Web services. We define methods for the creation of reputation information, its collection, and assessment that are robust in the face of a variety of attacks. Our framework (denoted RATEWeb) supports a cooperative model in which Web services share their experiences of the service providers with their peers through feedback ratings. The different ratings are aggregated to derive a service providerâ s reputation. This in turn is used to evaluate trust. For situations where rater feedbacks are scarce, we use statistical forecasting (particularly, a Hidden Markov Model) to ascertain trust. The approaches and techniques developed under the RATEWeb framework facilitate the optimal selection and/or composition of Web services based on service reputations. We conduct an extensive performance study (analytical and experimental) to assess the fairness and accuracy of the proposed techniques.