<?xml version="1.0" standalone="yes"?>
<Paper uid="A00-2018">
  <Title>A Maximum-Entropy-Inspired Parser *</Title>
  <Section position="2" start_page="0" end_page="0" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> We present a new parser for parsing down to Penn tree-bank style parse trees \[16\] that achieves 90.1~ average precision/recall for sentences of length &lt; 40, and 89.5% for sentences of length &lt; 100, when trained and tested on the previously established \[5,9,10,15,17\] &amp;quot;standard&amp;quot; sections of the Wall Street Journal tree-bank.</Paragraph>
    <Paragraph position="1"> This represents a 13% decrease in error rate over the best single-parser results on this corpus \[9\].</Paragraph>
    <Paragraph position="2"> Following \[5,10\], our parser is based upon a probabilistic generative model. That is, for all sentences s and MI parses 7r, the parser assigns a probability p(s, ~) = p(Tr), the equality holding when we restrict consideration to ~r whose yield * This research was supported in part by NSF grant LIS SBR 9720368. The author would like to thank Mark Johnson and all the rest of the Brown Laboratory for Linguistic Information Processing.</Paragraph>
    <Paragraph position="3"> is s. Then for any s the parser returns the parse 7r that maximizes this probability. That is, the parser implements the function arg. a= p( I = = arg maxTrp(lr).</Paragraph>
    <Paragraph position="4"> What fundamentally distinguishes probabilistic generative parsers is how they compute p(~r), and it is to that topic we turn next.</Paragraph>
  </Section>
class="xml-element"></Paper>