<?xml version="1.0" standalone="yes"?>
<Paper uid="W04-1207">
  <Title>Event-based Information Extraction for the biomedical domain: the Caderige project</Title>
  <Section position="6" start_page="47" end_page="48" type="concl">
    <SectionTitle>
6 Conclusion
</SectionTitle>
    <Paragraph position="0"> We have presented in this paper some results from the Caderige project. Two major issues are the development of a specific annotation editor for domain specialists and a set of machine learning and linguistic processing tools tuned for the biomedical domain.</Paragraph>
    <Paragraph position="1"> Current developments focus on the use of learning methods in the extraction process.</Paragraph>
    <Paragraph position="2"> These methods are introduced at different levels in the system architecture. A first use is  the acquisition of domain knowledge to enhance the extraction phase. A second use concerns a dynamic adaptation of existing modules during the analysis according to specific features in a text or to specific text genres.</Paragraph>
  </Section>
class="xml-element"></Paper>