A new TV recommendation algorithm based on interest quantification and item clustering

Recommender Systems(RSs) are software tools and techniques providing suggestions for items to be of use to a user. With the increasing development of Internet and explosion of information, recommender system has been an indispensable component in many applications. In this paper, a recommendation algorithm based on factorization model is proposed, which is applied to TV system. To quantize users' interest/preference to programs, a novel and rational notation, user interest index, is defined and helps improve recommendation effect. The vectorization of users and programs are derived from item clustering. Finally, we adopted top-K recommendation strategy, and evaluated the performance of our algorithm. According to experiment results, we found that the algorithm performs well on precision and recall rate.

[1]  Gediminas Adomavicius,et al.  Toward the next generation of recommender systems: a survey of the state-of-the-art and possible extensions , 2005, IEEE Transactions on Knowledge and Data Engineering.

[2]  Geoffrey E. Hinton,et al.  Restricted Boltzmann machines for collaborative filtering , 2007, ICML '07.

[3]  Thomas Hofmann,et al.  Latent semantic models for collaborative filtering , 2004, TOIS.

[4]  Sophie Ahrens,et al.  Recommender Systems , 2012 .

[5]  Lixin Gao,et al.  The impact of YouTube recommendation system on video views , 2010, IMC '10.

[6]  G. Takács,et al.  On the Gravity Recommendation System , 2007 .

[7]  Daniel Lemire,et al.  Slope One Predictors for Online Rating-Based Collaborative Filtering , 2007, SDM.

[8]  Fillia Makedon,et al.  Using singular value decomposition approximation for collaborative filtering , 2005, Seventh IEEE International Conference on E-Commerce Technology (CEC'05).

[9]  Yehuda Koren,et al.  Matrix Factorization Techniques for Recommender Systems , 2009, Computer.

[10]  Meng Chen,et al.  A study about the personalized recommendation system of TV program based on user behaviors , 2012, 2012 IEEE Fifth International Conference on Advanced Computational Intelligence (ICACI).

[11]  Yehuda Koren,et al.  The BellKor Solution to the Netflix Grand Prize , 2009 .

[12]  Yehuda Koren,et al.  Scalable Collaborative Filtering with Jointly Derived Neighborhood Interpolation Weights , 2007, Seventh IEEE International Conference on Data Mining (ICDM 2007).

[13]  Takao Terano,et al.  A TV Program Recommender Framework , 2013, KES.

[14]  Dean P. Foster,et al.  A Formal Statistical Approach to Collaborative Filtering , 1998 .