Comparison of techniques for time aware TV channel recommendation

With the increasing number of TV channels, it is more difficult for viewers to find their preferred TV channel. Thus, the recommender system for TV is needed. However, it has several difficulties. First, the viewer's preferred TV channel is different according to the temporal context. Moreover, the sparseness problem also occurs when we consider temporal context. Temporal context has been recognized as an important factor to consider in personalized recommender systems. A lot of time aware recommendation methods were proposed for these difficulties. In this paper, we survey and compare some techniques for time aware TV channel recommendation such as Singular Value Decomposition (SVD), traditional Matrix Factorization (MF), and Temporal Regularized Matrix Factorization (TRMF). We apply them for real-world data to analyze possible benefits of temporal context information for TV channel recommendation and compare the performance of each of them.

[1]  Keon-Myung Lee.  Adaptive Resource Scheduling for Workflows Considering Competence and Preference , 2004, KES.

[2]  Chun-Chia Lee,et al.  AIMED- A Personalized TV Recommendation System , 2006, EuroITV.

[3]  Sung-Bae Cho,et al.  A Hybrid Recommender System Based on AHP That Awares Contexts with Bayesian Networks for Smart TV , 2014, HAIS.

[4]  Bernd Ludwig,et al.  Matrix factorization techniques for context aware recommendation , 2011, RecSys '11.

[5]  Jee-Hyong Lee,et al.  A music recommendation system with a dynamic k-means clustering algorithm , 2007, Sixth International Conference on Machine Learning and Applications (ICMLA 2007).

[6]  Ee-Peng Lim,et al.  Modeling Temporal Adoptions Using Dynamic Matrix Factorization , 2013, 2013 IEEE 13th International Conference on Data Mining.

[7]  Fabio Bellifemine,et al.  User Modeling and Recommendation Techniques for Personalized Electronic Program Guides , 2004, Personalized Digital Television.

[8]  Yueh-Min Huang,et al.  Community-based program recommendation for the next generation electronic program guide , 2009, IEEE Transactions on Consumer Electronics.

[9]  Jee-Hyong Lee,et al.  Implementation of Ontology Based Context-Awareness Framework for Ubiquitous Environment , 2007, 2007 International Conference on Multimedia and Ubiquitous Engineering (MUE'07).

[10]  Nuria Oliver,et al.  Multiverse recommendation: n-dimensional tensor factorization for context-aware collaborative filtering , 2010, RecSys '10.

[11]  LeeWei-Po,et al.  A smart TV system with body-gesture control, tag-based rating and context-aware recommendation , 2014 .