<?xml version="1.0" standalone="yes"?>
<Paper uid="W05-0826">
  <Title>Combining Linguistic Data Views for Phrase-based SMT</Title>
  <Section position="5" start_page="147" end_page="147" type="concl">
    <SectionTitle>
4 Conclusions
</SectionTitle>
    <Paragraph position="0"> We have showed that it is possible to obtain better phrase-based translation models by utilizing alignments built on top of different linguistic data views. These models can be robustly combined, significantly outperforming all of their components in isolation. We leave for further work the experimentation of new data views such as word senses and semantic roles, as well as their natural porting and evolution from the alignment step to phrase extraction and decoding.</Paragraph>
    <Paragraph position="1"> 4Translation for 349 unknown words was found in the MCR.</Paragraph>
  </Section>
class="xml-element"></Paper>