STERS: A system for service trustworthiness evaluation and recommendation based on the trust network

Along with the rapid development of the Internet, more and more web services can be found and used. However, service trustworthiness is one of the most critical factors to keep the confidence of users in using them. In this paper, we propose a novel approach to evaluate the trustworthiness of web services using the Trust Network and give a mechanism for service recommendation in consideration of the users' different preference and perspectives. We adopt the Trust Network to estimate the important degree of subjective evidence from different resources and filter the false or malicious evidence. Then we calculate the trustworthiness of web services according to both of subjective and objective evidence. In addition to this, we put forward a preference template to recommend the proper service to the users according to the users' requirements. Finally, we develop a system called STERS to evaluate the service trustworthiness and recommend the service. Experiments conducted on a large-scale real-world dataset show that our method can effectively evaluate the trustworthiness of web services, which helps users to select and use them.