<?xml version="1.0" standalone="yes"?>
<Paper uid="W97-0312">
  <Title>Learning to Tag Multilingual Texts Through Observation</Title>
  <Section position="7" start_page="114" end_page="114" type="concl">
    <SectionTitle>
7 Summary
</SectionTitle>
    <Paragraph position="0"> RoboTag is a multilingual text extraction system that automatically learns to tag texts by observing its users. Decision trees are learned to predict where users begin and end tags. These are combined with a matching algorithm to produce complete tags.</Paragraph>
    <Paragraph position="1"> RohoTag is flexible in its ability to work with multiple languages. We have shown RoboTag performance to be competitive with hand-coded pattern-based systems in very different languages like English and Japanese.</Paragraph>
  </Section>
class="xml-element"></Paper>