<?xml version="1.0" standalone="yes"?>
<Paper uid="A92-1025">
  <Title>Joining Statistics with NLP for Text Categorization</Title>
  <Section position="9" start_page="184" end_page="184" type="concl">
    <SectionTitle>
7 Summary and Conclusion
</SectionTitle>
    <Paragraph position="0"> This paper has addressed the area where statistical and linguistic analysis come together with an application focus -- the assignment of categories to news stories, particularly the identification of topics and industries.</Paragraph>
    <Paragraph position="1"> The experiments reported here show that using statistical methods to acquire simple lexical patterns helps knowledge-based processing and leads to a substantial improvement in overall system performance. In addition, this method promises to ease the burden of hand-coding knowledge for each application, by automatically identifying the significant terms and combinations of terms to use knowledge-intensive NLP applications.</Paragraph>
  </Section>
class="xml-element"></Paper>