<?xml version="1.0" standalone="yes"?>
<Paper uid="W06-3802">
  <Title>Graph Based Semi-Supervised Approach for Information Extraction</Title>
  <Section position="9" start_page="1411" end_page="1411" type="concl">
    <SectionTitle>
7 Conclusion
</SectionTitle>
    <Paragraph position="0"> We introduce a general framework for semi-supervised learning based on mutual reinforcement in graphs. We construct generalized extraction patterns and deploy graph based mutual reinforcement to automatically identify the most informative patterns. We provide motivation for our approach from a graph theory and graph link analysis perspective. null We present experimental results supporting the applicability of the proposed approach to ACE Relation Detection and Characterization (RDC) task, demonstrating its applicability to hard information extraction problems. Our approach achieves a significant improvement over the base line supervised system especially when the number of labeled instances is small.</Paragraph>
  </Section>
class="xml-element"></Paper>