<?xml version="1.0" standalone="yes"?>
<Paper uid="W93-0305">
  <Title>HMM-based Part-of-Speech Tagging for Chinese Corpora</Title>
  <Section position="8" start_page="45" end_page="45" type="concl">
    <SectionTitle>
5 Concluding Remarks
</SectionTitle>
    <Paragraph position="0"> We have presented our initial effort for Chinese partdeg of-speech tagging using a first-order fully-connected hidden Markov model and Kupiec's concept of equivalence classes. The experimental results show that the tagging model is promising. We have also discussed our observations on some imperfections of the current model. In the near future, we will (1) use the whole UD corpus to further validate and verify the system, (2) try to implement a second-order HMM, (3) attempt to solve part of the unknown word tagging problem, (4) attempt to solve part of the compound word problem, (5) use heuristic rules for postprocessing the tagging output, (6) perform word identification and part-of-speech tagging concurrently, and (7) integrate the tagging HMM with the linguistic decoder of a Chinese speech recognition system.</Paragraph>
  </Section>
class="xml-element"></Paper>