<?xml version="1.0" standalone="yes"?>
<Paper uid="W02-1810">
  <Title>An Indexing Method Based on Sentences*</Title>
  <Section position="3" start_page="0" end_page="0" type="concl">
    <SectionTitle>
5. Conclusion
</SectionTitle>
    <Paragraph position="0"> This paper demonstrates how the method creates the index file and gives the sentences including keywords. It then shows an example that employs the method to discover the sentences containing the adjective-noun pairs and compute their mutual information. As it is shown, the method can effectively extract the sentences including specific words and make the real-time probabilistic computation possible. It is also easy to extend the algorithm to search for three or more specific words appearing in the same sentences or to obtain the intersection, union and difference of their sentence number sets.</Paragraph>
    <Paragraph position="1"> The method can be widely applied for many applications in Chinese information processing, such as information extraction, segmentation, tagging, parsing, semantic analysis, dictionary compilation and information retrieval. It is particularly fit for the situation of dealing with specific words and sentences in large-scale corpora and is a supporting tool for the researches of natural language processing.</Paragraph>
  </Section>
class="xml-element"></Paper>