<?xml version="1.0" standalone="yes"?>
<Paper uid="W94-0112">
  <Title>Bootstrapping Statistical Processing into a Rule-based Natural Language Parser</Title>
  <Section position="6" start_page="364" end_page="364" type="concl">
    <SectionTitle>
5 Conclusion
</SectionTitle>
    <Paragraph position="0"> We have described a &amp;quot;bootstrapping&amp;quot; method, which uses a broad-coverage, rule-based parser to compute probabilities while parsing a non-annotated corpus of NL text, and which incorporates those probabilities into the very same parser for use in analyzing new text. The results reported from an evaluation of this method show that it can significantly improve the speed and accuracy of the parser. A salient feature of this method is that it does not require the use of annotatod corpora or human-supervised training during the computation of the probabilities.</Paragraph>
  </Section>
class="xml-element"></Paper>