<?xml version="1.0" standalone="yes"?>
<Paper uid="W04-1018">
  <Title>Chinese Text Summarization Based on Thematic Area Detection</Title>
  <Section position="7" start_page="4" end_page="4" type="concl">
    <SectionTitle>
5 Conclusions
</SectionTitle>
    <Paragraph position="0"> In this paper, we have proposed a new summarization method based on thematic areas detection. By adopting a novel clustering analysis method, it can adaptively detect the different thematic areas in the document, and automatically determine K, the number of thematic areas. So the produced summary can both cover as many as different themes and reduce its redundancy obviously at the same time.</Paragraph>
    <Paragraph position="1"> For our experiment, we used three different parameters to evaluate the quality of the produced summaries in theme coverage and summarization redundancy. We achieved a better performance than the traditional non-thematic -areas-detection method in the proposed evaluation scheme. As a future work , we need the additional research for testing the proposed method on la rger-scale real corpora , and have the further comparison with earlier similar works such as MMR, etc. In addition, we'll improve our summarization system by considering the structure of thematic areas and user's requirement.</Paragraph>
  </Section>
class="xml-element"></Paper>