<?xml version="1.0" standalone="yes"?>
<Paper uid="P02-1065">
  <Title>Memory-Based Learning of Morphology with Stochastic Transducers</Title>
  <Section position="9" start_page="0" end_page="0" type="concl">
    <SectionTitle>
7 Conclusion
</SectionTitle>
    <Paragraph position="0"> We have presented some algorithms for the supervised learning of morphology using the EM algorithm applied to non-deterministic finite-state transducers. null We have shown that a novel Memory-based learning technique inspired by the Fisher kernel method produces high performance in a wide range of languages without the need for fine-tuning of parameters or language specific representations, and that it can account for some psycho-linguistic data. These techniques can also be applied to the unsupervised learning of morphology, as described in (Clark, 2001b).</Paragraph>
  </Section>
class="xml-element"></Paper>