Trust-aware Recommender System In Social Network

With the rapid development of the information technology and widespread use of Internet, the human race has entered the information society and the network era. People can make friends on social network, shop on the Internet conveniently and study via Internet whenever and wherever. However, the rapidly growing amount of available information and digital items on the web aggravates the severity of the information overload problem for online users. Due to the openness and anonymity of the Internet, the decision-making process becomes perplexing when one is exposed to excessive information. Although, recommender systems are highly successful and are frequently used in the real-world network, they have their limitations when handing challenges such as a large number of user, the explosive growth of information and data.In this paper, we propose a trust-aware recommender system to enhance the accuracy of recommendation. Trust, as the basis of human interaction, has been playing an important role in addressing information sharing, experience communication and electronic commerce. Trust, inherently has many facets indicating multiple and heterogeneous trust relationships between users. Trust-aware recommender systems employ trust information to filter extra information and improve the accuracy of the recommendation as well as users’ experience.1、Traditionally, research about trust assumes a single type of trust between users. However, trust, as a complex and abstract social concept, is inherently asymmetry and context dependence. We propose a novel approach by incorporating multi-faced trust relationships into traditional rating prediction algorithms to reliably estimate users’ trust strengths, then merge trust information, users’ preferences and items’ characteristics into classical recommender systems to improve users recommendation and address some of their challenges including the cold start problem and their weakness against the attack.2、Exploiting the context information such as time, relationships and user feedback information in social network, it presents a trust evolution social recommendation method based on user feedback. This method incorporates the temporal factors by introducing a time weight function, which models the decay of user interest. Moreover, the method considers the user feedback information to filter most fake profiles. In this case, the system will not use these users in recommendation.