<?xml version="1.0" standalone="yes"?>
<Paper uid="W93-0301">
  <Title>Robust Bilingual Word Alignment for Machine Aided Translation</Title>
  <Section position="5" start_page="5" end_page="7" type="concl">
    <SectionTitle>
4 Conclusions
</SectionTitle>
    <Paragraph position="0"> Compared with other word alignment algorithms (Brown et al., 1993; Gale and Church, 1991a), word_align does not require sentence alignment as input, and was shown to produce useful alignments for small and noisy corpora. Its robustness was achieved by modifying Brown et al.'s Model 2 to handle an initial &amp;quot;rough&amp;quot; alignment, reducing the number of parameters and introducing a dependency between alignments of adjacent words. Taking the output of char_align as input, word_align produces significantly better, word- null scanned by an OCR device.</Paragraph>
    <Paragraph position="1"> level, alignments on the kind of corpora that are typically available to translators. This improvement increased the rate of constructing bilingual terminology lexicons at AT&amp;T Language Line Services by a factor of 2-3. In addition, the alignments may also be helpful to developers of lexicons for machine translation systems. Word_align thus provides an example how a model such as Brown et al.'s Model 2, that was originally designed for research in statistical machine translation, can be modified to achieve practical, though less ambitious, goals in the near term.</Paragraph>
  </Section>
class="xml-element"></Paper>