Trust measurement using fuzzy theory and trade protocol recommendation based on trust level in trusted auction system

The explosive growth of the internet-based transactions requires not only a secure payment system but also an appropriate trust measuring methodology and secure transaction protocols to guarantee the minimal risk for the transacting entities involved in specific transactions. Especially, in internet auction systems where either buyers or sellers or both can be more than one in one transaction, providing those systems that make sure no one transacting entity takes a major risk becomes critical. In this paper, an improved trust measuring method using a relationship-based internet community for an auction system is proposed. The proposed system incorporates fuzzy set and calculation concepts to help build trust matrices and models, which is used to measure the level of risk involved in a specific auction trade concerned. Also, to optimize the auction trade process in terms of cost and time, the proposed system recommends a differentiated trade protocol according to the risk level involved in each auction trade. To test the appropriateness of the proposed trusted auction system, a prototype system has been developed under a Windows-NT environment.