<?xml version="1.0" standalone="yes"?>
<Paper uid="P06-1089">
  <Title>Guessing Parts-of-Speech of Unknown Words Using Global Information</Title>
  <Section position="7" start_page="711" end_page="711" type="concl">
    <SectionTitle>
5 Conclusion
</SectionTitle>
    <Paragraph position="0"> In this paper, we presented a method for guessing parts-of-speech of unknown words using global information as well as local information. The method models a whole document by considering interactions between POS tags of unknown words with the same lexical form. Parameters of the model are estimated from training data using Gibbs sampling. Experimental results showed that the method improves accuracies of POS guessing of unknown words especially for Chinese and Japanese. We also applied the method to semi-supervised learning, but the results were not consistent and there is some room for improvement.</Paragraph>
  </Section>
class="xml-element"></Paper>