A collaborative recommender system based on probabilistic inference from fuzzy observations

The problem of building recommender systems has attracted considerable attention in recent years. The objective of this paper is to automatically suggest and rank a list of new items to a user based on the past voting patterns of other users with similar tastes. The proposed model can be considered as a Soft Computing-based collaborative recommender system. The combination of Bayesian networks, which enables an intuitive representation of the mechanisms that govern the relationships between the users, and the Fuzzy Set Theory, enabling us to represent ambiguity or vagueness in the description of the ratings, improves the accuracy of the system. © 2008 Elsevier B.V. All rights reserved.