<?xml version="1.0" standalone="yes"?>
<Paper uid="I05-2026">
  <Title>Lexical Chains and Sliding Locality Windows in Content-based Text Similarity Detection</Title>
  <Section position="7" start_page="153" end_page="153" type="concl">
    <SectionTitle>
6 Conclusion
</SectionTitle>
    <Paragraph position="0"> We have presented a system that extends previous work on lexical chains to content similarity detection. This system employs lexical chains and sliding locality windows, and evaluates similarity using the cosine of n-grams of lexical chains and tf*idf weighted keywords.</Paragraph>
    <Paragraph position="1"> The results indicate that lexical chains are effective for detecting content similarity between pairs of chapters corresponding to the same original in a corpus of parallel translations.</Paragraph>
  </Section>
class="xml-element"></Paper>