Recommender systems make information filtering for user by predicting user’s preference to items. Collaborative filtering is the most popular technique in implementing a recommender system. Association rule mining is a powerful data mining method to search for interesting relationships between items by finding the items frequently appeared together in a transaction database. In this paper, we apply quantitative association rules to mining the relationships between items, and then utilize the relationships between items to alleviate the data sparsity problem in the neighborhood-based algorithms. The proposed method considers not only similarities between users, but also similarities between items. The experimental results on two publicly available datasets show that our algorithm outperforms the conventional Pearson method and adjusted cosine method.