Effective Web content recommendation based on consumer behavior modeling

Web surfing has become an important activity for many consumers. We propose a web recommender that models user behavior by constructing a know ledge base using temporal web access patterns as input. We apply fuzzy logic to represent real-life temp oral concepts and requested resources of periodic pattern-based web access activities. Th e fuzzy representation is used to construct a know ledge base of the user's web access behaviors, which is then used to provide timely and personalized recommendations to the user, possibly on their portable devices. Experiments conducted to e valuate the performance of the proposed approach have shown very promising results