A Novel Automatic Text Summarization Study Based on Term Co-Occurrence

Abstract

As the amount of textual information available grows rapidly, automatic text summarization methods are becoming increasingly important. Based on the subject information from term co-occurrence graph and linkage information of different subjects, a novel automatic summarization algorithm is proposed in this paper. This algorithm can get better summarization and can be adaptable to different document style. And it also can pick up subject information, whose significance was evaluated in accordance to the rules presented in this paper. Besides, it can dynamically decide the summary size. The validity of the algorithm has been tested, showing that the novel automatic text summarization algorithm tallies with author's intentional subjects and is information-redundancy-free