<?xml version="1.0" standalone="yes"?>
<Paper uid="C90-2032">
  <Title>Sentence disambiguation by document preference sets oriented</Title>
  <Section position="1" start_page="0" end_page="0" type="abstr">
    <SectionTitle>
231, JAPAN
Abstract:
</SectionTitle>
    <Paragraph position="0"> This paper proposes document oriented preference sets(DoPS) for the disambiguation of the dependency structure of sentences. The I)oPS system extracts preference knowledge from a target document or other documents automatically. Sentence ambiguities can be resolved by using domain targeted preference knowledge without using complicated large knowledgebases. Implementation and empirical results are described for the cmalysis of dependency structures of Japanese patent claim sentences.</Paragraph>
    <Paragraph position="1"> To solve this problem, we introduce Document oriented Preference Sets(DoPS). The concept of DoPS is that to determine the most appropriate preference knowledge, preference knowledge be segregated into several domains, for example, language domain, field domain, and sentence domain, each of which has a different execution priority. By using the segregated preference knowledge in the fixed order, the most plausible interpretation can be obtained more rapidly and more accurately.</Paragraph>
  </Section>
class="xml-element"></Paper>