<?xml version="1.0" standalone="yes"?>
<Paper uid="W98-0318">
  <Title>Automatic Disambiguation of Discourse Particles</Title>
  <Section position="6" start_page="111" end_page="112" type="concl">
    <SectionTitle>
5 Results and Conclusion
</SectionTitle>
    <Paragraph position="0"> To determine in how far the system relying only on the two types of information described, syntactic position and the previous dialogue act, can automatically disambiguate occurrences of discourse particles, a test was designed on the basis of natural authentic dialogues. These, however, were typed into the system in order to be able to analyze the contribution of the implementation most clearly; the results are therefore not obscured from errors by the speech recognizer. null Good results were achieved regarding the  test-sentences with altogether 75 discourse particles. In comparison with a doublechecked hand-tagged version, 83% were automatically assigned the correct discourse function. These results show that automatic speech processing systems can indeed disambiguate discourse particle occurrences on the basis of extremely reduced linguistic knowledge. Consequently it is not the case that automatic speech processing systems cannot identify the relevant information. Thus, discourse particles do not constitute such a chaotic domain after all; already very simple and easily obtainable linguistic information on their occurrences in dialogues can lead to their automatic classification. A system which could make use of prosody, for instance, may get even better results. This opens up the way to employing discourse particles in automatic speech processing systems, for instance, as keywords regarding propositional information, to infer dialogue structure, or in order to control the information flow and to support speech management tasks, which all can support the aims of automatic speech processing systems.</Paragraph>
  </Section>
class="xml-element"></Paper>