<?xml version="1.0" standalone="yes"?>
<Paper uid="W00-1101">
  <Title>Adapting a synonym database to specific domains</Title>
  <Section position="6" start_page="4" end_page="4" type="concl">
    <SectionTitle>
4 Conclusions
</SectionTitle>
    <Paragraph position="0"> There is a need for automatically or semi-automatically adapting NLP components to specific domain, if such components are to be effectively used in IR applications without involving labor-intensive manual adaptation. A key part of adapting NLP components to specific domains is the adaptation of their lexical and terminological resources. It may often be the case that a consistent section of a general purpose terminological resource is irrelevant to a specific domain, thus involving an unnecessary amount of ambiguity that affects both the accuracy and efficiency of the overall NLP component. In this paper we have proposed a method for adapting a general purpose synonym database to a specific domain.</Paragraph>
    <Paragraph position="1"> Evaluating the performance of the proposed pruning method is not a straightforward task, since there are no other results available on a similar task, to the best of our knowledge. However, a comparison between the results of manual and automatic pruning provides some useful hints. In particular: * The discrepancy between the evaluation of human operators shows that the task is elusive even for humans (the value of the agreement evaluation statistic n for our human evaluators was 0.5); * however, the correlation between the level of confidence of human evaluations and scores assigned by the automatic pruning procedure shows that the automatic pruning algorithm captures some significant aspect of the problem.</Paragraph>
    <Paragraph position="2"> Although there is probably room for improving the automatic pruning performance, the preliminary results show that the current approach is pointing in the right direction.</Paragraph>
  </Section>
class="xml-element"></Paper>