<?xml version="1.0" standalone="yes"?>
<Paper uid="C02-1075">
  <Title>A Novel Disambiguation Method For Unification-Based Grammars Using Probabilistic Context-Free Approximations</Title>
  <Section position="8" start_page="89" end_page="89" type="concl">
    <SectionTitle>
7 Conclusion
</SectionTitle>
    <Paragraph position="0"> This paper concerns the problem of how to disambiguate the readings of sentences, analyzed by a given UBG.</Paragraph>
    <Paragraph position="1"> We presented a novel approach to disambiguation for UBGs, shifting the responsibility to simpler CF models, obtained by the approximation of the UBG. In contrast to earlier approaches to disambiguation for UBGs, our approach can be effectively applied in practice, enables unsupervised training on free text corpora, as well as efficient disambiguation, and is mathematically clean.</Paragraph>
    <Paragraph position="2"> We showed that our novel approach is feasible for a mid-size UBG of English. Evaluation of an unsupervised trained model achieved a precision of 88% on an exact match task.</Paragraph>
  </Section>
class="xml-element"></Paper>