Study on the Recommendation Trust Model Based on Encouragement and Punishment

In the open distributed environment, the knowledge of belief is absent for the entity with which we will contact, so the recommendation trust is very important for the distributed system. In this paper a recommendation trust model is proposed based on encouragement and punishment. In the model, the trust value of any entity consists of trading reputation value and recommendation reputation value, the recommendation factor is affected by multi parameters and can adjust dynamically with the recommend result. When the recommender supports the real information then the value of the recommendation trust will increase otherwise decrease the value.