<?xml version="1.0" standalone="yes"?>
<Paper uid="I05-2045">
  <Title>Unsupervised Feature Selection for Relation Extraction</Title>
  <Section position="5" start_page="266" end_page="266" type="concl">
    <SectionTitle>
4 Conclusion and Future work
</SectionTitle>
    <Paragraph position="0"> In this paper, we presented an unsupervised approach for relation extraction from corpus. The advantages of the proposed approach includes that it doesn't need any manual labelling of the relation instances, it can identify an important feature subset and the number of the context clusters automatically, and it can avoid extracting those common words as characterization of relations.</Paragraph>
  </Section>
class="xml-element"></Paper>