<?xml version="1.0" standalone="yes"?>
<Paper uid="W97-0210">
  <Title>m m Investigating Complementary Methods for Verb Sense Pruning</Title>
  <Section position="7" start_page="63" end_page="63" type="concl">
    <SectionTitle>
5 Discussion
</SectionTitle>
    <Paragraph position="0"> Our method for using detailed knowledge about verb subcategorizations and alternations to prune verb senses is domain independent. It also prunes senses without loss of correctness. By intersecting the resulting sense sets with the output of our cluster-based method, verb senses can be pruned further.</Paragraph>
    <Paragraph position="1"> In using the clustering method's output, we make two further assumptions. Previous work has shown that within a given discourse (Gale et al., 1992), or with respect to a given collocation (Yarowsky, 1993), a word appears in only one sense. By extrapolation, we will assume that words appear in only one sense within a homogeneous corpus, 4 except for certain high frequency verbs or for semantically empty support verbs. We will assign this predominant sense to all non-disambignated occurrences of a verb. While this provides a reasonable default, the resulting semantic tag has to be considered provisional, and validated independently. Also, we currently assume that words placed in the same group will share relatively few links (connecting pairs of competing senses) in WordNet. This is supported by our initial experiments, but is an issue we will continue to investigate. Above we gave some preliminary evaluation results; we plan to carry out a more complete evaluation of our system by continuing to use the hand-tagged (with WordNet senses) Brown corpus (Miller et al., 1993) as the initial evaluation standard. Each stage will be separately measured, as well as their combined effectiveness in pruning senses. We anticipate that the use of multiple methods to investigate sense pruning will lead to more robust results. In addition, we believe that the two methods can be interleaved in the following manner: Both methods rely tOt a few predominant senses, that can perhaps be disambigu&amp;ted using syntactic constraints as we discuss below.</Paragraph>
    <Paragraph position="2"> on recognizing features of the local syntactic context of a verb occurrence; the look-up method uses the local syntactic context to identify the likely subcategorization pattern while the automatic classification method uses the local syntactic context to extract marker words. The look-up method can tag distinct tokens of the same verb with distinct senses if the subcategorization patterns are distinct and correlate with distinct senses. The automatic classification method could be extended to classify sense sets, using as its input corpus the output of the syntactic constraints look-up method, where verb tokens have been tagged with a subset of the full collection of senses. In principle, this would make it possible to use the automatic classification method on a more heterogeneous corpus, i.e., where the same verb occurs frequently with two distinct senses.</Paragraph>
  </Section>
class="xml-element"></Paper>