A Computational Trust Model in C2C E-Commerce Environment

In C2C e-commerce environment it is possible and indeed quite common to deal with unknown sellers, whose reliability are unknown. Reputation system is a popular technique currently used for providing a global quality score of sellers to consumers. However, such global information is far enough for clients to choose the most qualified products. In the real word, we always ask our trusted friends for some recommended information or trust the information recommended by the clients who have similar preference in recommender systems. Aiming at providing some more useful trust information to consumers, we propose a computational trust model based on the social network which is constructed by the similarity of users in the C2C e-commerce environment, and use a local trust metric for propagating trust in the constructed network. Experiments show that our model can produce a personalized recommended trust value for a specific user, which is more accuracy than a global value in some cases.