<?xml version="1.0" standalone="yes"?>
<Paper uid="W06-0906">
  <Title>Extending TimeML with Typical Durations of Events</Title>
  <Section position="7" start_page="43" end_page="44" type="concl">
    <SectionTitle>
5 Conclusion
</SectionTitle>
    <Paragraph position="0"> In this paper we have demonstrated how to extend TimeML with typical durations of events.</Paragraph>
    <Paragraph position="1"> We can see that the extension is very straightforward. Other interesting temporal information can be extracted or learned. For example, for each event class, we can generate its own mean and widths graphs, and learn their durations separately from other classes, which may capture different duration characteristics associated with each event class.</Paragraph>
  </Section>
class="xml-element"></Paper>