<?xml version="1.0" standalone="yes"?>
<Paper uid="P05-2009">
  <Title>Learning Meronyms from Biomedical Text</Title>
  <Section position="8" start_page="53" end_page="53" type="concl">
    <SectionTitle>
6 Conclusions
</SectionTitle>
    <Paragraph position="0"> The PartEx system is capable of fully automated learning of meronyms between semantically typed terms, from the experimental corpus. With simulated pattern pruning, it achieves a recall of 0.73 and a precision of 0.58. In contrast to earlier work, these results were achieved without manual labelling of the corpus, and without direct manual selection of high performance patterns. Although the cost of this automation is lower results than the earlier work, failure analyses provide insights into the algorithm and scope for its further improvement.</Paragraph>
    <Paragraph position="1"> Current work includes: automated pattern pruning, extending pattern context and generalisation; incorporating deeper analyses of the text, such as semantic labelling (c.f. Girju (2003)) and the use of dependency structures; investigating the r^ole of term recognition in relation discovery; measures for evaluating new relation discovery; extraction of putative sub-relations of meronymy. Work to scale the algorithm to larger corpora is also under way, in recognition of the fact that the corpus used was small, highly regularised, and unusually rich in meronyms.</Paragraph>
  </Section>
class="xml-element"></Paper>