<?xml version="1.0" standalone="yes"?>
<Paper uid="W04-3227">
  <Title>Phrase Pair Rescoring with Term Weightings for Statistical Machine Translation</Title>
  <Section position="8" start_page="31" end_page="31" type="concl">
    <SectionTitle>
7 Conclusion and Discussion
</SectionTitle>
    <Paragraph position="0"> In this work, we proposed a way of using term weight based models in a vector space as additional evidences for translation quality, and integrated the model into an existing phrase-based statistical machine translation system. The model shows significant improvements when using it to score a manual dictionary as well as when using different phrase transducers or a combination of all available translation information. Additional experiments also confirmed the effectiveness of the proposed model in terms of of improved Mean Reciprocal Rank of good translations. null Our future work is to explore alternatives such as the reranking work in (Collins, 2002) and include more knowledge such as syntax information in rescoring the phrase translation pairs.</Paragraph>
  </Section>
class="xml-element"></Paper>