<?xml version="1.0" standalone="yes"?>
<Paper uid="P98-2148">
  <Title>A Stochastic Language Model using Dependency and Its Improvement by Word Clustering</Title>
  <Section position="7" start_page="902" end_page="902" type="concl">
    <SectionTitle>
6 Conclusion
</SectionTitle>
    <Paragraph position="0"> In this paper we have presented dependency models for Japanese based on the attribute of bunsetsu.</Paragraph>
    <Paragraph position="1"> They are the first fully stochastic dependency models for Japanese which describes from character sequence to syntactic tree. Next we have proposed a word clustering method, an extension of deleted interpolation technique, which has been proven to be efficient in terms of improvement of the predictive power. Finally we have discussed parsers based on our model which demonstrated a remarkable improvement in parsing accuracy by our word-clustering method.</Paragraph>
  </Section>
class="xml-element"></Paper>