ABSTRACT
Lack of trust has been identified as a major obstacle to the growth of online transactions. The overall goal of this research is to develop two kinds of computational trust and reputation models to facilitate reliable cooperation within virtual enterprises. Firstly, this paper presents an improved trust model, which combines five trust decisive factors to compute the trustworthiness of partners. Through simulating experiment, this model's performance is validated. Second, in order to address the scarcity of personalization in the above model, a novel recommendation based trust model is proposed which utilizes the theory of collaborative filtering technology and social networks, and includes two indexes of trust scenario similarity and propensity to trust similarity. Finally, comparing our model with Beth model based on eBay's data, the result shows the feasibility and effectiveness of this approach.
