Improving Temporal Recommendation Accuracy and Diversity via Long and Short-Term Preference Transfer and Fusion Models

Temporal factor plays an important role in products and services recommended process. It is necessary to combine temporal factors with effective methods to improve recommendation performance. In this paper, we present a novel approach to improve personalized recommendation performance with changing user preferences based on temporal dataset. In the approach, we take consideration of different influence of long and short-term user preferences and construct a preference transfer model based on our enhanced Hidden Markov Model. Then we accomplish preference fusion by adopting our Long and Short Term Graph, a graph model modified from Session-based Temporal Graph, to recommend unknown items. Finally, the experimental results show that our approach achieves important improvements compared to some existing approaches in performance.

[1]  David M. Pennock,et al.  Probabilistic Models for Unified Collaborative and Content-Based Recommendation in Sparse-Data Environments , 2001, UAI.

[2]  Hui Xiong,et al.  Enhancing Collaborative Filtering by User Interest Expansion via Personalized Ranking , 2012, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[3]  Daniel Dajun Zeng,et al.  Collaborative filtering in social tagging systems based on joint item-tag recommendations , 2010, CIKM.

[4]  Duen-Ren Liu,et al.  Hybrid Recommendation Approaches: Collaborative Filtering via Valuable Content Information , 2005, Proceedings of the 38th Annual Hawaii International Conference on System Sciences.

[5]  John Riedl,et al.  Item-based collaborative filtering recommendation algorithms , 2001, WWW '01.

[6]  Evaggelia Pitoura,et al.  Managing contextual preferences , 2011, Inf. Syst..

[7]  Yehuda Koren,et al.  Collaborative filtering with temporal dynamics , 2009, KDD.

[8]  Jimeng Sun,et al.  Temporal recommendation on graphs via long- and short-term preference fusion , 2010, KDD.

[9]  Param Vir Singh,et al.  A Hidden Markov Model for Collaborative Filtering , 2010, MIS Q..

[10]  Cherif Chiraz Latiri,et al.  LC-mine: a framework for frequent subgraph mining with local consistency techniques , 2014, Knowledge and Information Systems.

[11]  Mahdi Jalili,et al.  A probabilistic model to resolve diversity–accuracy challenge of recommendation systems , 2015, Knowledge and Information Systems.

[12]  Alexander Felfernig,et al.  Content-based recommendation techniques for requirements engineering , 2014, 2014 IEEE 1st International Workshop on Artificial Intelligence for Requirements Engineering (AIRE).

[13]  Nuanwan Soonthornphisaj,et al.  Hybrid Recommendation: Combining Content-Based Prediction and Collaborative Filtering , 2003, IDEAL.

[14]  Robin Burke,et al.  Adapting Recommendations to Contextual Changes Using Hierarchical Hidden Markov Models , 2015, RecSys.

[15]  Licia Capra,et al.  Temporal collaborative filtering with adaptive neighbourhoods , 2009, SIGIR.