<?xml version="1.0" standalone="yes"?>
<Paper uid="E99-1010">
  <Title>An Efficient Method for Determining Bilingual Word Classes</Title>
  <Section position="8" start_page="187" end_page="187" type="concl">
    <SectionTitle>
6 Summary and future works
</SectionTitle>
    <Paragraph position="0"> By applying a maximum-likelihood approach to the joint probability of a parallel corpus we obtained an optimization criterion for bilingual word classes which is very similar to the one used in monolingual maximum-likelihood word clustering.</Paragraph>
    <Paragraph position="1"> For optimization we used the exchange algorithm.</Paragraph>
    <Paragraph position="2"> The obtained word classes give a low translation lexicon perplexity and improve the quality of sta- null Proceedings of EACL '99 tistical machine translation.</Paragraph>
    <Paragraph position="3"> We expect improvements in translation quality by allowing that words occur in more than one class and by performing a hierarchical clustering. Acknowledgements This work has been partialIy supported by the European Community under the ESPRIT project number 30268 (EuTrans).</Paragraph>
  </Section>
class="xml-element"></Paper>