<?xml version="1.0" standalone="yes"?>
<Paper uid="C04-1069">
  <Title>Document Re-ranking Based on Automatically Acquired Key Terms in Chinese Information Retrieval</Title>
  <Section position="7" start_page="133" end_page="133" type="concl">
    <SectionTitle>
6 Conclusion
</SectionTitle>
    <Paragraph position="0"> In this paper, we proposed a new method to improve the precision of top N initial ranking documents in Chinese IR. We try to find proper and important long terms in queries and documents, then we make use of these information to reweight the similarity between queries and documents and finally reorder the top M (M&gt;N) documents by their new similarities with query. Our experiences based on bigram as indexing and word as indexing both show that our method can improve the performance of Chinese IR by 10%-11% at top 10 documents measure level and 2%-5% at top 100 documents document measure level. For the further work, we will try to improve the quality of Global Key Terms and Local Key Terms, and we will apply our method to English IR and other languages IR systems.</Paragraph>
  </Section>
class="xml-element"></Paper>