A digital television navigator
MULTIMEDIA '00 Proceedings of the eighth ACM international conference on Multimedia
| Hi-index | 0.00 |
An intelligent television content personalization system has been designed and presented in this paper. It can detect program information by stream analysis, identify different users through face recognition, learn user's preference by intelligent algorithm, and offer dynamic recommendations to help every user quickly find, locate and organize their favorite programs. TV program is automatically rated based on the ratio of user's total watching time to the content's real duration; system learns user preference by calculating and comparing the rating of each program. The function and efficiency of the system is tested by an experiment, showing that this system can adapt user's preference and give correct recommendations.