<?xml version="1.0" standalone="yes"?>
<Paper uid="E06-1041">
  <Title>Structuring Knowledge for Reference Generation: A Clustering Algorithm</Title>
  <Section position="8" start_page="326" end_page="327" type="concl">
    <SectionTitle>
6 Conclusions and future work
</SectionTitle>
    <Paragraph position="0"> This paper attempted to achieve a dual goal. First, we highlighted a number of scenarios in which the performance of a GRE algorithm can be enhanced by an initial step which identifies clusters of entities or properties. Second, we describedan algorithmwhich takesas input a set of objects and returns a set of clusters based on a calculation of their perceived proximity. The definition of perceived proximity seeks to take into account some of the principles of human perceptual and conceptual organisation.</Paragraph>
    <Paragraph position="1"> In current work, the algorithm is being applied to  twoproblemsinGRE,namely,thegenerationofspatial references involving collective predicates (e.g. gathered), and the identification of the available perspectives or conceptual covers, under which referents may be described.</Paragraph>
  </Section>
class="xml-element"></Paper>