In this paper, we propose a multi-channel dissemination system with a clustering mechanism and a presentation technique for time-series on-line new articles on the Internet. We describe a detecting technique for articles whose topics are the same. These articles are called follow-up articles. In addition, we describe a calculating method for confidence level and scoop level assigned to news articles using temporal information. Furthermore we describe a a prototypee system called Carthage based on our proposed method. The system reconstructs all of the articles to follow-up article groups and distribute the article group with several user interfaces.
展开▼