<?xml version="1.0" standalone="yes"?>
<Paper uid="P03-1002">
  <Title>Using Predicate-Argument Structures for Information Extraction</Title>
  <Section position="13" start_page="0" end_page="0" type="concl">
    <SectionTitle>
5 Conclusion
</SectionTitle>
    <Paragraph position="0"> This paper reports on a novel inductive learning method for identifying predicate argument structures in text. The proposed approach achieves over 88% F-measure for the problem of identifying argument constituents, and over 83% accuracy for the task of assigning roles to pre-identified argument constituents. Because predicate lexical information is used for less than 5% of the branching decisions, the generated classifier scales better than the statistical method from (Gildea and Palmer, 2002) to unknown predicates. This way of identifying predicate argument structures is a central piece of an IE paradigm easily customizable to new domains.</Paragraph>
    <Paragraph position="1"> The performance degradation of this paradigm when compared to IE systems based on hand-crafted patterns is only 10%.</Paragraph>
  </Section>
class="xml-element"></Paper>