Subjective Trust Evaluation Model Based on Preferences

Abstract

In the open network environment, in the face of vast amounts of resources and services, due to the fraud and the existence of unreliable service, the user increased choice, at the same time how to identify and select efficient and secure resources or services. So the trust evaluation has become one of the focuses of current research. Because the trust relationship between entities is highly subjective, uncertainty, ambiguity, drawing on the trust model of human sociology, according to preference entities interested in, considering the various attributes and characteristics, adopts the fuzzy comprehensive evaluation method to trust the entity evaluation. In order to increase the recommended credibility and reliability of recommended entities, the introduction of services similarity algorithm in the Web semantic network, the weighted harmonic and recommendation trust value, in order to improve the reliability of recommended entities, and finally considering the direct trust and recommendation trust of entities, accordingly conclude both meet the current needs of the requesting entity safe and reliable service entities. Finally, simulation experiments performed to verify the validity of model credibility assessment.