Research On Trust Model And Mechanism In P2P E-Commerce

With the development of network technology and spread of personal computers, P2P technology has developing rapidly in recent years. Applying P2P technology which is dynamic, openness and anonymity to e-commerce makes e-commerce become much more flexible. However at the same time, P2P technology makes security and trust issues in e-commerce become worse. The building of reputation evaluation system and trust mechanism can effectively solve trust problems and help users to establish a reliable, security trust relationship. However, most existing P2P trust models are based on common P2P applications, there are only a few P2P trust models that setting e-commerce as the application environment, and the ability of these trust models to defense malicious attacks is relatively low, so this paper do following research work on reputation evaluation, trust model and mechanism in e-commerce:1. Based on objects involved in C2C e-commerce, a model of users'satisfaction is proposed, a transaction satisfaction evaluation indicator system is established by building users'concerns and indicators of satisfaction; AHP(Analytic Hierarchy Process) is used to determine the weight of these indicators, the integrated transaction satisfaction of users which also known as the user's reputation value is weighted average. Comparing to existing reputation evaluation indicators which are very simple, our indicator system takes a full consideration of sellers, logistics, payment, websites, goods and other related objects, so the evaluation is much more comprehensive. Comparing to "Good, Medium and Bad Review", transaction satisfaction which is the weighted average based on several indicators can reach the evaluation which is more objective and closer to the truth, and make the trust value using transaction satisfaction as main input more reliable.2. This paper proposes a new method of calculating cumulative reputation. This method considers the influence from transaction amount in every transaction to the increment of reputation, the impact of time to reputation is also considered. Cumulating reputation in stages is proposed in this method, the original cumulative reputation is weakened by using time functions at the beginning of every new stage. Follow-up evaluating is proposed by this method, which can enhance the true extent of cumulative reputation. This paper suggests dividing cumulative reputation into long-term and short-term, the reputation information of the seller is intuitively and vividly showed by demonstrating each graphically.3. This paper establishes a new trust evaluation model for e-commerce, transaction satisfaction is used as the main input of trust calculation in this model, direct trust and cumulative reputation is integrated to support users of making trading decisions. The method of implementing e-commerce trust mechanism in P2P networks is proposed, including storage mechanism of trust data, query, calculate and update mechanisms of trust value, and ensure data security and complaint mechanisms.