<?xml version="1.0" standalone="yes"?>
<Paper uid="H92-1023">
  <Title>Decision Tree Models Applied to the Labeling of Text with Parts-of-Speech</Title>
  <Section position="8" start_page="119" end_page="120" type="concl">
    <SectionTitle>
7. Conclusions
</SectionTitle>
    <Paragraph position="0"> In two experiments we have seen how decision trees provide modest improvements over HMM's for the problem of labeling unrestricted text with parts-of-speech. In examining the errors made by the models which incorporate the decision tree marginals, we find that the errors may be attributed to two primary problems: bad ques-</Paragraph>
  </Section>
class="xml-element"></Paper>