Identifying fake feedback in cloud trust management systems using feedback evaluation component and Bayesian game model

摘要

Cloud computing trust management has become an important subject in recent years. Trust management is a difficult and complicated work in cloud computing due to features such as distributed, dynamic and non-transparent environment. Since most trust management frameworks predict trust values based on received feedbacks, the validation and authenticity of feedback are highly necessary. In this paper, two new methods are presented to identify fake feedbacks. One is feedback evaluation component and the other is Bayesian game model. The feedback evaluation component is used to examine the received feedback and identify its probable fake identity. Our results reveal that the feedback evaluation component can correctly identify and rectify fake feedbacks. Bayesian game model is presented to detect malicious users and prevent their feedbacks. Simulation results coincided well with analytical results, indicating that our Bayesian game model can correctly recognize malicious user. Received feedbacks from malicious user are identified as fake feedbacks. Eventually, two new methods were compared.