Integrating Fuzzy Partitional Clustering and Collaborative Filtering for Web Page Prediction

摘要

Web page prediction is a popular personalized service on the Web and has attracted much research attention. Collaborative filtering (CF) technique identifies other users that have shown similar preferences to the web pages and recommends what they have liked to that user. Traditional CF techniques require explicit and extra user participation for providing his/her preference to the pages. In addition, the computational cost of computing the nearest neighbors increases linearly with the number of users and items which makes them unsuitable for online processing. To address these problems, we utilize an implicit indicator which measures the time spent on a page. In particular, we propose a cluster-based collaborative filtering algorithm based on the fuzzy set theory, which is designed to better address the sharp boundary problem of discretizing the viewing time. The experimental results show that the proposed method can achieve a better performance than the traditional click-only CF approach.