<?xml version="1.0" standalone="yes"?>
<Paper uid="J96-4006">
  <Title>Integrating General-purpose and Corpus-based Verb Classification</Title>
  <Section position="4" start_page="566" end_page="566" type="concl">
    <SectionTitle>
4. Conclusions
</SectionTitle>
    <Paragraph position="0"> The results illustrated in this paper are very interesting, though not conclusive. Verbs are vessels for human creativity in language communication, and so much is left to further studies. We discovered thematic features that are apparently more &amp;quot;basic&amp;quot; than others, with respect to a given semantic domain (cognition) and a given sublanguage (RSD). We could specify features that were described at a very general level in Word-Net, and detect semantic restrictions specific to the sublanguage, not accounted for in WordNet. These results suggest that, with appropriate customization, it is still possible to exploit the information in general-purpose on-line thesauri that would be otherwise almost unusable in real NLP applications. As proposed in this paper, an appropriate process of lexical tuning can significantly reduce the overgenerality (excessive ambiguity) and underspecificity (weak constraints on verb argument structures) that is typical of general-purpose resources.</Paragraph>
  </Section>
class="xml-element"></Paper>