<?xml version="1.0" standalone="yes"?>
<Paper uid="P91-1026">
  <Title>AUTOMATIC NOUN CLASSIFICATION BY USING JAPANESE-ENGLISH WORD PAIRS*</Title>
  <Section position="6" start_page="206" end_page="206" type="concl">
    <SectionTitle>
5. CONCLUSION
</SectionTitle>
    <Paragraph position="0"> Using word pairs of Japanese and English to distinguish among senses of the same verb, we have shown that using word pairs to classify nouns is better than not using word pairs, when only a few nouns are classified. However, this experiment did not succeed for a sufficient number of nouns for two reasons. One is that the raw co-occurrent frequency is used to calculate the semantic distance. The other is that the sample size is too small. It is thus necessary to resolve the following issues to make the classification result more accurate.</Paragraph>
    <Paragraph position="1"> (1)to develop a means of using the frequency normalized by expected word pairs.</Paragraph>
    <Paragraph position="2"> (2)to estimate an adequate sample size.</Paragraph>
    <Paragraph position="3"> In this experiment, we acquired word pairs and semantic relations from our database. However, they are made by hand.</Paragraph>
    <Paragraph position="4"> It is also preferable to develop a method of automatically acquiring them from the bilingual text database.</Paragraph>
    <Paragraph position="5"> Moreover, we want to apply the hierarchically classified result to the translated word selection problem in Machine translation.</Paragraph>
  </Section>
class="xml-element"></Paper>