Improve Recommendation by Trust Mechanism: From Perspective of Web 2.0

Web 2.0 and collaborative filtering recommendation technologies have been widely exploited to unleash the creation appetency of people for sharing contents with others and receiving customized services in an information overload time. In this paper, by studying the growing process of online communities we propose a trust mechanism as well as related algorithms to improve recommendation, especially for the new user cold start problem. Through the experiments under the scale-free network environment, our work shows a more flexible and natural property to enhances user´s acceptance of the Web 2.0 application.