<?xml version="1.0" standalone="yes"?>
<Paper uid="P98-2176">
  <Title>Learning Correlations between Linguistic Indicators and Semantic Constraints: Reuse of Context-Dependent Descriptions of Entities</Title>
  <Section position="8" start_page="1077" end_page="1077" type="concl">
    <SectionTitle>
7 Conclusion
</SectionTitle>
    <Paragraph position="0"> In this paper, we showed that context and other linguistic indicators correlate with the choice of a particular noun phrase to describe an entity. Using machine learning techniques from a very large corpus, we automatically extracted a large set of rules that predict the choice of a description out of an entity profile. We showed that high-precision automatic prediction of an appropriate description in a specific context is possible.</Paragraph>
  </Section>
class="xml-element"></Paper>