An Effective Two-Phase Collaborative Filtering Algorithm for Recommender Systems Bo-Wen Wang1, Ja-Hwung Su1, Chin-Yuan Hsiao2 and Vincent S. Tseng1* 1 Department of Computer Science and Information Engineering, National Cheng Kung University, Tainan, Taiwan, R.O.C. 2 Industrial Technology Research Institute of Taiwan, R.O.C *tsengsm@mail.ncku.edu.tw ABSTRACT. A recommender system refers to a set of machine learning methods that learn from a user’s behavior to deal with the problem of information overload. Although traditional collaborative filtering has been shown to be effective in predicting a user’s preferences, it suffers from a data sparsity problem. To alleviate such sparsity problem, in the present study, an innovative collaborative filtering recommender that decomposes the prediction procedure into two phases is proposed. In the first phase, the user’s unknown ratings are imputed as the initial ratings to provide information for the second prediction phase. Experimental evaluation results show the effectiveness of the proposed method, especially for situations with very sparse data. Keywords: recommender system, collaborative filtering, item-based recommender, user-based recommender, two-phase recommender 1