<?xml version="1.0" standalone="yes"?>
<Paper uid="W05-0617">
  <Title>Morphology Induction From Term Clusters</Title>
  <Section position="7" start_page="189" end_page="189" type="concl">
    <SectionTitle>
5 Conclusion
</SectionTitle>
    <Paragraph position="0"> We have shown that automatically computed term clusters can form the basis of an effective unsupervised morphology induction system. Such clusters tend to group terms by part of speech, greatly simplifying the search for syntactically significant affixes. Furthermore, the learned affixation patterns are not just orthographic features or morphological conflation sets, but cluster-to-cluster transformation rules. We exploit this in the construction of morphological automata able to analyze previously unseen wordforms.</Paragraph>
    <Paragraph position="1"> We have not exhausted the sources of evidence implicit in this framework, and we expect that attending to features such as transform frequency will lead to further improvements. Our approach may also benefit from the kinds of broad-context semantic filters proposed elsewhere. Finally, we hope to use the cluster assignments suggested by the morphological rules in refining the original cluster assignments, particularly of low-frequency words.</Paragraph>
  </Section>
class="xml-element"></Paper>