RESEARCH ON TRUST PREDICTION MODEL FOR SELECTING WEB SERVICES BASED ON MULTIPLE DECISION FACTORS

Abstract

With the widespread application of web services, trust-based service selection has become a significant requirement from a requester's point of view, and trust prediction mechanism has become a determining factor for any given service's success. But the dynamic nature of trust creates the biggest challenge in measuring a trust value and make trust predictions. In this paper, a new trust prediction model is proposed based on multiple decision factors. Firstly, the proposed model integrates multiple complementary decision factors to reflect complexity and uncertainty of trust relationship. In addition to including traditional experience-based trust factor and feedback-based trust factor, this model innovatively integrates QoS-based trust factor into global trust evaluation, which makes the model have a better rationality and a higher practicability than the existing approaches. Meanwhile, the proposed model applies the theory of OWA (Ordered Weighted Averaging) operator to assign the classification weights to these decision factors, which makes the model exhibit a strong adaptability in handling various dynamic behaviors of service providers. Simulation results show that, compared to the existing approaches, the proposed model has remarkable enhancements in the accuracy of trust prediction.