An Improved User-model-based Collaborative Filtering Algorithm ⋆

Collaborative filtering is an algorithm successfully and widely used in recommender system. However, it suffers from data sparsity, recommendation accuracy and system scalability problems. This paper proposes an improved user model for collaborative filtering to explore a solution to these problems. The ratings are firstly been normalized by decoupling normalization method, and then a nonlinear forgetting function is introduced to assign the ratings different time weights to mimic the users’ interest drift. In similarity computation, an effective weighting factor is added to the Pearson correlation similarity computation to get more accurate neighbor users. The algorithm is tested on MovieLens dataset and the comparative experiment shows that the algorithm proposed in this paper can provide a better performance.

[1]  George M. Giaglis,et al.  A hybrid approach for improving predictive accuracy of collaborative filtering algorithms , 2007, User Modeling and User-Adapted Interaction.

[2]  Ken Lang,et al.  NewsWeeder: Learning to Filter Netnews , 1995, ICML.

[3]  Panagiotis Symeonidis,et al.  Collaborative recommender systems: Combining effectiveness and efficiency , 2008, Expert Syst. Appl..

[4]  Luo Si,et al.  Collaborative filtering with decoupled models for preferences and ratings , 2003, CIKM '03.

[5]  John Riedl,et al.  GroupLens: an open architecture for collaborative filtering of netnews , 1994, CSCW '94.

[6]  Michael R. Lyu,et al.  Effective missing data prediction for collaborative filtering , 2007, SIGIR.

[7]  Cao Xianbin,et al.  Research on Lineal Gradual Forgetting Collaborative Filtering Algorithm , 2007 .

[8]  Duen-Ren Liu,et al.  Product recommendation approaches: Collaborative filtering via customer lifetime value and customer demands , 2008, Expert Syst. Appl..

[9]  Derek G. Bridge,et al.  Collaborative Recommending using Formal Concept Analysis , 2006, Knowl. Based Syst..

[10]  Qian Wang,et al.  Collaborative filtering recommendation algorithm based on hybrid user model , 2010, 2010 Seventh International Conference on Fuzzy Systems and Knowledge Discovery.

[11]  Dennis McLeod,et al.  Yoda: An Accurate and Scalable Web-Based Recommendation System , 2001, CoopIS.

[12]  SongJie Gong,et al.  An Efficient Collaborative Filtering Algorithm with Item Hierarchy , 2008, 2008 Second International Symposium on Intelligent Information Technology Application.

[13]  Christoph Froschl,et al.  User Modeling and User Profiling in Adaptive E-learning Systems , 2005 .

[14]  Li Lei,et al.  An Enhanced Collaborative Filtering Algorithm Based on Time Weight , 2009, 2009 International Symposium on Information Engineering and Electronic Commerce.

[15]  Hidekazu Tsuji,et al.  情報システム,エレクトロニック・コマース An improved hybrid recommender system using multi-based clustering method (特集 若手研究者) , 2009 .

[16]  SongJie Gong,et al.  Mining User Interest Change for Improving Collaborative Filtering , 2008, 2008 Second International Symposium on Intelligent Information Technology Application.