摘要

Trust relationship is one of the most key and complex concepts in P2P network societies, so its quantitative model inevitably includes multiple decision-making factors. In this paper, using theory of entropy-based information fusion, a novel dynamic trust model is proposed innovatively, in which multiple complementary decision factors are incorporated to reflect complexity and uncertainty of the trust relationship. Meanwhile, the weights of these decision factors are assigned by theory of entropy-based multiple information fusion, which overcomes the shortage of traditional methods, in which the weights are assigned by subjective manners, and makes the model has a better rationality and a higher practicability. Simulated results show that, compared to the existing approaches, the proposed model exhibits a remarkable enhancements in the accuracy of trust prediction.