A word-of-mouth information recommender system considering information reliability and user preferences

In this paper, with the aim of facilitating information gathering on the Internet, we propose an information recommender system based on a word-of-mouth model that passes obtained information rapidly from one user to another while paying attention to information reliability and user preferences. The system uses a method for improving information reliability by treating user reliability and information reliability as equivalent and using an evaluation method employed on Internet auctions. It also uses a method that accommodates user preferences by changing the user that initiates communication (passes information) by using a personalized word-of-mouth agent that records user preferences. We report on very interesting phenomena revealed by our experimental results.