<?xml version="1.0" standalone="yes"?>
<Paper uid="P01-1006">
  <Title>Evaluation tool for rule-based anaphora resolution methods</Title>
  <Section position="5" start_page="0" end_page="0" type="concl">
    <SectionTitle>
4 Conclusion
</SectionTitle>
    <Paragraph position="0"> We believe that the evaluation workbench for anaphora resolution proposed in this paper 7In Tables 1 and 2, only pronouns that are treated as anaphoric and hence tried to be resolved by the three methods are included. Therefore, pronouns in first and second person singular and plural and demonstratives do not appear as part of the number of pronouns.</Paragraph>
    <Paragraph position="1"> alleviates a long-standing weakness in the area of anaphora resolution: the inability to fairly and consistently compare anaphora resolution algorithms due not only to the difference of evaluation data used, but also to the diversity of pre-processing tools employed by each system.</Paragraph>
    <Paragraph position="2"> In addition to providing a common ground for comparison, our evaluation environment ensures that there is fairness in terms of comparing approaches that operate at the same level of automation: formerly it has not been possible to establish a correct comparative picture due to the fact that while some approaches have been tested in a fully automatic mode, others have benefited from post-edited input or from a pre- (or manually) tagged corpus. Finally, the evaluation workbench is very helpful in analysing the data used for evaluation by providing insightful statistics.</Paragraph>
  </Section>
class="xml-element"></Paper>