<?xml version="1.0" standalone="yes"?>
<Paper uid="W05-1208">
  <Title>A Probabilistic Setting and Lexical Cooccurrence Model for Textual Entailment</Title>
  <Section position="9" start_page="47" end_page="47" type="concl">
    <SectionTitle>
6 Conclusions
</SectionTitle>
    <Paragraph position="0"> This paper proposes a generative probabilistic setting that formalizes the notion of probabilistic textual entailment, which is based on the conditional probability that a hypothesis is true given the text.</Paragraph>
    <Paragraph position="1"> This probabilistic setting provided the necessary grounding for a concrete probabilistic model of lexical entailment that is based on document co-occurrence statistics in a bag of words representation. Although the illustrated lexical system is relatively simple, as it doesn't rely on syntactic or other deeper analysis, it nevertheless achieved encouraging results. The results suggest that such a probabilistic framework is a promising basis for improved implementations incorporating deeper types of knowledge and a common test-bed for more sophisticated models.</Paragraph>
  </Section>
class="xml-element"></Paper>