<?xml version="1.0" standalone="yes"?>
<Paper uid="P98-2158">
  <Title>A DP based Search Algorithm for Statistical Machine Translation</Title>
  <Section position="5" start_page="965" end_page="965" type="concl">
    <SectionTitle>
4 Conclusion and Future Work
</SectionTitle>
    <Paragraph position="0"> In this paper, we have presented a new search algorithm for statistical machine translation. First experiments prove its applicability to realistic and complex tasks such as spontaneously spoken dialogs.</Paragraph>
    <Paragraph position="1"> Several improvements to our algorithm are planned, the most important one being the implementation of pruning methods (Ney et al., 1992). Pruning methods have already been used successfully in machine translation (Tillmann et al., 1997a). The first question to be answered in this context is how to make two different hypotheses H1 and/-/2 comparable: Even if they cover the same number of source string words, they might cover different words, especially words that are not equally difficult to translate, which corresponds to higher or lower translation probability estimates. To cope with this problem, we will introduce a heuristic for the estimation of the cost of translating the remaining source words.</Paragraph>
    <Paragraph position="2"> This is similar to the heuristics in A'-search.</Paragraph>
    <Paragraph position="3"> (Vogel et al., 1996) report better perplexity results on the Verbmobil Corpus with their HMM-based alignment model in comparison to Model 2 of (Brown et al., 1993). For such a model, however, the new interpretation of the alignments becomes essential: We cannot adopt the estimates for the alignment probabilities p(ili', I). Instead, we have to re-calculate them as inverted alignments. This will provide estimates for the probabilities P(JlJ', J).</Paragraph>
    <Paragraph position="4"> The most important advantage of the HMM-based alignment models for our approach is the fact, that they do not depend on the unknown target string length I.</Paragraph>
    <Paragraph position="5"> Acknowledgement. This work was partly supported by the German Federal Ministry of Education, Science, Research and Technology under the Contract Number 01 IV 601 A (Verbmobil).</Paragraph>
  </Section>
class="xml-element"></Paper>