Dynamic Trust Model Based On Extended Subjective Logic

Trust is a kind of believe or expectation, it is a subjective concept of mutual and asymmetry. That is to say, nodes exist trust evaluation on both sides of each other, and they aren’t always the same. In J?sang’s model, not only above problem but also the impact of event weight to the trust evaluation is not considered. Besides, the trust evaluation is updated dynamically and continuously as the time going on and the process of interactions, but it has not been reflected in J?sang’s model. Thus the final trust evaluation is not accurate, the malicious fraud cannot be prevented effectively, which cause the success rate of interaction is low. Besides, most of traditional trust search using flooding method searching trust path. This algorithm has low efficiency and high computational cost, and it also cannot solve the issue of overlapping of friends-circle.To solve this issues, a new dynamic trust model based on extended subjective logic(DTM-ESL) is proposed. In DTM-ESL, the event weight and the mutuality of trust are fully considered, the original one-way trust relationship is extended to the two-way trust relationship, and three trust strategies about the relative trust evaluation between two nodes are given based on event weight; the influence of event weight to the node’s recommended qualifications is fully considered, and a new trust search algorithm is proposed, the path dependence and trust-circle caused by the overlapping of friends-circle are solved effectively by the strategies of dual-threshold screening and set added; the discounting operator and consensus operator in J?sang’s trust model are improved, and the trust renewal is designed based on the event weight and the decay of trust evaluation, including the time decay of trust evaluation, trust reward and punishment. Finally, the viability and effectiveness of new model are verified by simulation experiments.