<?xml version="1.0" standalone="yes"?>
<Paper uid="W04-2009">
  <Title>Recovering Coherent Intepretations Using Semantic Integration of Partial Parses</Title>
  <Section position="10" start_page="8" end_page="8" type="concl">
    <SectionTitle>
9 Conclusion
</SectionTitle>
    <Paragraph position="0"> The analysis task found within a learning model exhibits all the canonical problems that motivate robust parsing algorithms: extra/ungrammaticality, restarts, agreement issues etc. Because of these reasons, the language learning setting seems like a good testing ground for robust analysis algorithms.</Paragraph>
    <Paragraph position="1"> One such algorithm has been described in this paper. Thealgorithm is unique because of its heavy reliance on semantics. It uses knowledge at all phases of its analysis, from when it recognizes constructions using a semantic chunking approach, to when it merges the chunks' common semantic structures.</Paragraph>
    <Paragraph position="2"> This paper also takes a small step toward defining a gradient notion of semantic analysis. Not all analyses of an utterance are created equal. One simple approach to comparing semantic interpretations is by measuring how completely they specify their schemas and frames. This is how the semantic density metric works.</Paragraph>
    <Paragraph position="3"> More important than the particulars of these algorithms, however, is the idea that a learning model and an analysis model should be tightly coupled.</Paragraph>
    <Paragraph position="4"> Such a model makes it possible for a language understanding system to learn new language from experience. If such an idea can come to fruition, this would be the most robust language analysis algorithm of all.</Paragraph>
  </Section>
class="xml-element"></Paper>