<?xml version="1.0" standalone="yes"?>
<Paper uid="P91-1019">
  <Title>SUBJECT-DEPENDENT CO-OCCURRENCE AND WORD SENSE DISAMBIGUATION</Title>
  <Section position="8" start_page="151" end_page="151" type="concl">
    <SectionTitle>
CONCLUSION
</SectionTitle>
    <Paragraph position="0"> Although the words in the LDOCE definitions constitute a small text (almost one million words, compared with the mega-texts used in other co-occurrence studies), the unique feature of subject codes which can be used to distinguish many definitions, and LDOCE's small control vocabulary (2,187 words) make it a useful corpus for obtaining co-occurrence data.</Paragraph>
    <Paragraph position="1"> The development of techniques for information retrieval and word-sense disambiguation based on these subject-dependent co-occurrence neighborhoods is very promising indeed.</Paragraph>
  </Section>
class="xml-element"></Paper>