<?xml version="1.0" standalone="yes"?>
<Paper uid="W99-0907">
  <Title>Detecting Sub-Topic Correspondence through Bipartite Term Clustering</Title>
  <Section position="6" start_page="49" end_page="49" type="ackno">
    <SectionTitle>
Acknowledgements
</SectionTitle>
    <Paragraph position="0"> We thank Dekang Lin for the use of his Automatically Generated Thesaurus. This research was supported by ISRAEL SCIENCE  cost-based scheme for cx values between 0.53 to 0.73. Coupling connections are indicated as straight lines. The gray areas indicate coupled subsets of lower granularity level, which were formed for 0.53 &lt; cx &lt; 0.68 (I), 0.53 ~ ct __. 0.55 (11) and cx = 0.53 (111). Terms that swapped their membership for different o~ values appear, whenever possible, in the intersection of the appropriate subsets; otherwise, they are marked with asterisk. Terms that for some cx value were not included in any coupled subset are in Italics.</Paragraph>
  </Section>
class="xml-element"></Paper>