A Combination of Content-based Filtering and Item- based Collaborative Filtering Using Association Rules

In this paper we present a hybrid approach that tries to alleviate the main item-based collaborative filtering (CF) drawback, the sparsity and the first-rater problem. By combining the contents of items into the item-based CF to find similar items and use the combined similarity to generate predictions. We describe and evaluate how similarity attributes can be discovered and exploited using association rules mining and an approach to provides a way to introduce content information into item-based CF. The experiments show that our approach achieve better prediction accuracy than traditional itembased CF framework, especially for the sparsity problem.