Temporal Ordering of News Corpus
Hiroshi Uejima, Takao Miura · 2004
In this investigation, we propose a new mechanism to give timestamps to a collection of news corpus without any timestamps. Here we learn temporal data in advance to extract temporal feature by means of incre- mental clustering and then we estimate most likely timestamps to each news text. In this work, we examine TDT2 corpus and we show how well our approach works by some experiments.