Limiting Sybil Attacks on Bayesian Trust Models in Open SOA Environments

Next generation SOA systems promise to enable an "Internet of Services" - an open environment, in which every participant is not only free to offer and consume services, but also to provide central components of the ecosystem, such as marketplaces and community portals. In such an environment with no or only very little central control, malicious users may try to gain profits by providing low quality services. An approach to solve this issue is to introduce a trust or reputation system, enabling market participants to identify trustworthy parties. An open market does not have access restrictions to the market, hence such systems are likely to be exposed to Sybil attacks. In this paper, we provide a novel approach for aggregating recommendations in the face of Sybil attacks, which allow a single party to try to multiply the influence of its recommendation(s) by creating a high number of seemingly independent entities.