<?xml version="1.0" standalone="yes"?>
<Paper uid="C02-1166">
  <Title>An approach based on multilingual thesauri and model combination for bilingual lexicon extraction</Title>
  <Section position="7" start_page="0" end_page="0" type="concl">
    <SectionTitle>
Conclusion
</SectionTitle>
    <Paragraph position="0"> We have shown in this paper how to optimally combine different models derived from different resources for bilingual lexicon extraction from comparable corpora. We have proven that such a combination significantly (by 30%) improves the results over the models alone. We have also presented different models based on a multilingual thesaurus, and have obtained the best results with the model integrating hierarchical information. Lastly we have proposed different ways to enrich existing thesauri with new terms discovered in parallel corpora. Future work should focus on terminology extraction from comparable corpora.</Paragraph>
  </Section>
class="xml-element"></Paper>