A Trusted Service Selection Method Based on User's Personality Feature and Service Recommendation

Abstract

In view of the fact that the QoS-based service selection method in the current service selection method is less concerned with the personality attribute characteristics of the service requesters and the service selection method based on collaborative filtering, the service providers', Based on the characteristics of the personality of the service requester, this paper describes the user correlation by defining the similarity and domain relevance of the user, and the calculation method of the recommended credibility is given by using the credible measurement theory. Using the analytic hierarchy process (AHP) to determine the weight of each correlation factor, this paper proposes a credible service selection model based on collaborative filtering service selection trust model (SSTM). The simulation results show that the model can effectively improve the efficiency of service selection and resist the attack of malicious feedback. There are two major innovations as following: Firstly, to make a introduction of user relevance to reflect the degree of close between two users (service requester) under the network environment; to apply the user’s personality attribute characteristics to the service provider reputation value during the prediction, to improve the accuracy of service selection by reducing the size of service providers. Secondly, combining user relevance and recommendation credibility organically, using AHP to determine the weight of relevant factors in the service selection index system so that we can make the reputation of the predicted service provider more reliable and effectively resist the malicious user feedback.