Decision-making based on risk attribute in decentralized system

Reputation reporting systems have emerged as an important risk management mechanism in online trading communities. A trust referral system´s task is to use past ratings in order to calculate reliable reputation estimates for sellers. In this context, a reputation estimate acts as a predictor of a seller´s future service quality. Online rating systems have been widely adopted by online trading communities to ban “not good” service providers and prompt them to provide “good” services. However, the performance of the online rating systems is easily compromised by various unfair ratings, how to mitigate the influence of the unfair ratings remains a vital issue to improve the reliability of reputation systems. A variety of approaches have been proposed to cope with this issue without considering risk attribute of evaluator, so there are some disadvantages such as complexity of formula and huge computation. In this paper, a scheme based on buyer´s risk attribute is proposed in order to effectively play down the substantive computation.