<?xml version="1.0" standalone="yes"?>
<Paper uid="P04-3034">
  <Title>Fragments and Text Categorization</Title>
  <Section position="8" start_page="91" end_page="91" type="concl">
    <SectionTitle>
7 Conclusion
</SectionTitle>
    <Paragraph position="0"> We have introduced two methods - skip-tail and fragments - utilized for document categorization which are based on splitting documents into its subparts. We observed that both methods resulted in significant increase of accuracy. We also tested a method which exploited only the most confident fragments. However, this did not result in any accuracy increase. However, use of the most confident fragments for text summarization should also be checked.</Paragraph>
  </Section>
class="xml-element"></Paper>