<?xml version="1.0" standalone="yes"?>
<Paper uid="A97-1050">
  <Title>Semi-Automatic Acquisition of Domain-Specific Translation Lexicons</Title>
  <Section position="6" start_page="345" end_page="345" type="concl">
    <SectionTitle>
5 Conclusion
</SectionTitle>
    <Paragraph position="0"> In this paper, we have investigated the application of SABLE, a turn-key translation lexicon construction system for non-technical users, to the problem of identifying domain-specific word translations given domain-specific corpora of limited size. Evaluated on a very small (400,000 word) corpus, the system shows real promise as a method of processing small domain-specific corpora in order to propose candidate single-word translations: once likely general usage terms are automatically filtered out, the system obtains precision up to 89% at levels of recall very conservatively estimated in the range of 30-40% on domain-specific terms.</Paragraph>
    <Paragraph position="1"> Of the proposed entries not immediately suitable for inclusion in a translation lexicon, many represent part-of-speech divergences (of the protect/protdgg variety) and a smaller number incomplete entries (of the immddiatement/right variety) that would nonetheless be helpful if used as the basis for a bilingual concordance search -- for example, a search for French segments containing immddiatemeut in the vicinity of English segments containing right would most likely yield up the obvious correspondence between immgdiatement and right away. Going beyond single-word correspondences, however, is a priority for future work.</Paragraph>
  </Section>
class="xml-element"></Paper>