Automatic Summarization for Chinese Text Based on Combined Words Recognition and Paragraph Clustering

With the tremendous amount of information available electronically, there is an increasing requirement for automatic text summarization systems. An extractive summarization method is represented. The weight of a Chinese word/phrase is computed based on its frequency, part of speech, position and length. The weight of a Chinese sentence is computed by its content, position, length and cue words in it. The adjacent paragraphs are clustered into same cluster or different clusters according to their similarity. The experiment results show that the proposed algorithm has a significantly better performance compared with the traditional automated summarization algorithms based on TF-ISF method.