<?xml version="1.0" standalone="yes"?>
<Paper uid="I05-2034">
  <Title>Probabilistic Models for Korean Morphological Analysis</Title>
  <Section position="8" start_page="201" end_page="201" type="concl">
    <SectionTitle>
5 Conclusion
</SectionTitle>
    <Paragraph position="0"> We have presented and described the new probabilistic models used in our Korean morphological analyzer ProKOMA. The previous systems depend on manually constructed linguistic knowledge such as morpheme dictionary, morphosyntactic rules, and morphological rules. The system, however, requires no manual labor because all the information can be automatically acquired by the POS tagged corpora. We also showed that the system is portable and flexible by the experiments on three different corpora.</Paragraph>
    <Paragraph position="1"> The previous systems take morpheme as a processing unit, but we take three kinds of processing units (e.g. Eojeol, morpheme, and syllable). According to the experiments, we can know that the Eojeol-unit analysis contributes efficiency and accuracy, and the syllable-unit analysis is robust in the unknown word problem and also contributes accuracy. Finally, the system achieved comparable performances with the previous systems.</Paragraph>
  </Section>
class="xml-element"></Paper>