论文信息 - Channel Predictions for Cable TV Users by Modified Speed Algorithms

Channel Predictions for Cable TV Users by Modified Speed Algorithms

Cable TV users in China have quite large market share. Research on channel switch prediction plays a very important role in further understanding of user behavior. In this paper, we proposed a SPEED UPDATE algorithm to predict channels for cable TV users using the watching history, which improves 15% accuracy in comparison with the traditional SPEED algorithm. Furthermore, we developed a S-Markov algorithm which has the similar performance with the SPEED UPDATE algorithm and simplifies the computation process. We provided numerical simulations to illustrate the efficiency of our algorithms.

Fulian Yin | Jianhong Wu | Pei Su | Xuesong Bai

[1] Mamun Bin Ibne Reaz,et al. SPEED: An Inhabitant Activity Prediction Algorithm for Smart Homes , 2012, IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans.

[2] Diane J. Cook,et al. Online Sequential Prediction via Incremental Parsing: The Active LeZi Algorithm , 2007, IEEE Intelligent Systems.

[3] David Salomon,et al. Data Compression: The Complete Reference , 2006 .

[4] Brian D. Davison,et al. Predicting Sequences of User Actions , 1998 .

[5] Samy Bengio,et al. Scheduled Sampling for Sequence Prediction with Recurrent Neural Networks , 2015, NIPS.

[6] Longbing Cao,et al. Mining Partially-Ordered Sequential Rules Common to Multiple Sequences , 2015, IEEE Trans. Knowl. Data Eng..

[7] J. Ross Quinlan,et al. C4.5: Programs for Machine Learning , 1992 .

[8] Sheetal Kalyani,et al. A Modified PPM Algorithm for Online Sequence Prediction Using Short Data Records , 2015, IEEE Communications Letters.

[9] Sajal K. Das,et al. LeZi-Update: An Information-Theoretic Framework for Personal Mobility Tracking in PCS Networks , 2002, Wirel. Networks.

[10] Ben J. M. Smeets,et al. Towards understanding and improving escape probabilities in PPM , 1997, Proceedings DCC '97. Data Compression Conference.