<?xml version="1.0" standalone="yes"?>
<Paper uid="P98-2125">
  <Title>Identifying Syntactic Role of Antecedent in Korean Relative Clause Using Corpus and Thesaurus Information</Title>
  <Section position="7" start_page="761" end_page="761" type="concl">
    <SectionTitle>
6 Concluding Remarks
</SectionTitle>
    <Paragraph position="0"> This paper describes an approach for syntactic role determination between an antecedent and a verb in relative clause for semantic analysis. This method consists of two phases. In the first phase, the system extracts conceptual patterns and syntactic role distribution of antecedents from a large corpus. In the second phase, the system applies the extracted conceptual patterns as knowledge in determining correct syntactic relations for structural disambiguation and semantic analysis in MT system for CG generation.</Paragraph>
    <Paragraph position="1"> Unlike previous research that calculates statistical information at a lexical level for every pair of words, which may require a lot of space to store resulting patterns, we represent those co-occurrence patterns with concept types of Kadokawa thesaurus. The problematic concept types are filtered out by the type generalization procedure. We used a corpus of 6 million words for conceptual pattern extraction.</Paragraph>
    <Paragraph position="2"> Our method can cope with the general scope of texts. In the experiment evaluation, the proposed method showed a high accuracy rate of 90.4% in identifying the syntactic role of antecedents. null The method described in this paper can be used in resolving syntactic role of antecedents in relative clauses of other free word order languages, and can also be used in generating selectional restrictions of case frames of verbs.</Paragraph>
  </Section>
class="xml-element"></Paper>