<?xml version="1.0" standalone="yes"?>
<Paper uid="W02-1112">
  <Title>Using the WordNet Hierarchy for Associative Anaphora Resolution</Title>
  <Section position="6" start_page="0" end_page="0" type="concl">
    <SectionTitle>
5 Conclusions and Further Work
</SectionTitle>
    <Paragraph position="0"> Our intention in this paper has been to explore how we might automatically derive from a corpus a set of axioms that can be used in conjunction with an existing anaphor resolution mechanism; in particular, it is likely that in conjunction with an approach based on saliency, the axioms could serve as one additional factor to be included in computing the relative likelihood of competing antecedents.</Paragraph>
    <Paragraph position="1"> The preliminary results presented above do not in themselves make a strong case for the usefulness of the technique presented in this paper. However, they do suggest a number of possibilities for further work. In particular, we have begun to consider the following.</Paragraph>
    <Paragraph position="2"> First, we can make use of word sense disambiguation to reduce the negative consequences of generalising to synsets. Second, we intend to explore whether it is possible to determine an appropriate level of generalisation based on the class of the anaphor and antecedent. Third, there is scope for building on existing work on learning selectional preferences for WSD and the resolution of syntactic ambiguity; we suspect that, in particular, the work on learning class-to-class selectional preferences by (Agirre and Martinez, 2001) may be useful here.</Paragraph>
    <Paragraph position="3"> We are also looking for better ways to assess the results of using the axioms. Two directions here are clear. First, so far we have only a relatively small number of hand-annotated examples, from a single source. Increasing the number of examples will let us investigate questions like whether different choices of parameters are appropriate to different classes of anaphor. Second, it should be possible to refine the evaluation metrics: it is likely that even without looking at the effect of different filters in the context of a particular anaphora resolution system, we could provide a more meaningful analysis of their probable impact.</Paragraph>
    <Paragraph position="4"> In our current work, we have not explored the possibility of using information about associations that is explicitly encoded in existing machine-acessible ontologies. WordNet, for example, actually encodes meronym relationships. Our reason for not relying on this information in the first place was the limited set of relationships that were encoded, and the fact that associative relationships were encoded far less reliably than the hypernym relationship. However, it would be interesting to compare the results that could be obtained by using the ontology as a source for associative axioms with those that could be achieved by automatically deriving axioms from the data.</Paragraph>
    <Paragraph position="5"> Another direction we have not explored is the complementary information about anaphora resolution that derives from explicit statements of association: in line with the Gricean maxims, the author's decision to use an expression such as the leg of the okapi may constitute evidence that there is more than one previously mentioned entity in the context that may have legs. This information might be used, for example, to rule out an otherwise most likely antecedent. null In conclusion, we have shown in this paper how associative axioms can be derived automatically from a corpus, and we have explored how these axioms can be used to filter the set of candidate antecedents for instances of associative anaphora. Our initial evaluation of the impact of using these filters suggests that they are of limited value; yet the intuition that generalisations of this kind should be useful remains strong, and so our next steps are to find ways of refining and improving the approach.</Paragraph>
  </Section>
class="xml-element"></Paper>