Information filtering based on user behavior analysis and best match text retrieval
SIGIR '94 Proceedings of the 17th annual international ACM SIGIR conference on Research and development in information retrieval
Automatic parsing and indexing of news video
Multimedia Systems
Personalized Video for Contents Delivery Network
ASIAN '02 Proceedings of the7th Asian Computing Science Conference on Advances in Computing Science: Internet Computing and Modeling, Grid Computing, Peer-to-Peer Computing, and Cluster
A Rule-Based Scheme to Make Personal Digests from Video Program Meta Data
DEXA '01 Proceedings of the 12th International Conference on Database and Expert Systems Applications
Implicit news recommendation based on user interest models and multimodal content analysis
Proceedings of the 3rd international workshop on Automated information extraction in media production
Multimedia Tools and Applications
| Hi-index | 0.00 |
In this paper, we propose a new recommendation system for a TV news with automatic recomposition. For the time consuming browsing of the TV news articles, we propose three modes of presentation, the digest mode, the relaxed mode, and the normal mode, where each presentation length is different. To make these presentation, TV news articles are decomposed, analyzed, and stored in the database scene by scene. Then, the system selects desired items and synthesizes these scenes into a presentation based on a user's profile. For the profile of the user, we use a keyword vector and a category vector of news articles. The system is designed so that user's control to the system becomes minimum. Therefore, a user only plays, skips, plays previous, and rewinds news articles in the system as same as an ordinary TV. However, different from an ordinary TV, the system collects user's behavior while he uses the system. Based on this information, the system updates the user's profile. We also show preliminary experimental results.