<?xml version="1.0" standalone="yes"?>
<Paper uid="P98-2231">
  <Title>Structural Disambiguation Based on Reliable Estimation of Strength of Association Haodong Wu Eduardo de Paiva Alves</Title>
  <Section position="5" start_page="1420" end_page="1421" type="concl">
    <SectionTitle>
4 Concluding Remarks
</SectionTitle>
    <Paragraph position="0"> We proposed a class-based method that selects classes to be used to estimate the strength of association for word co-occurrences. The classes selected by our method can be used to estimate various types of strength of association in different applications. The method differs from other class-based methods in that it allows identification of a reliable and specific class for each co-occurrence in consideration and can deal with date sparseness problem more efficiently. It  overcame the shortcomings from other methods: overgeneralization and employment of unreliable class co-occurrences.</Paragraph>
    <Paragraph position="1"> We applied our method to two structural disambiguation experiments. In both experiments the performance is significantly better than those of others.</Paragraph>
  </Section>
class="xml-element"></Paper>