<?xml version="1.0" standalone="yes"?>
<Paper uid="P06-3007">
  <Title>Sydney, July 2006. c(c)2006 Association for Computational Linguistics Investigations on Event-Based Summarization</Title>
  <Section position="9" start_page="41" end_page="41" type="concl">
    <SectionTitle>
7 Conclusion
</SectionTitle>
    <Paragraph position="0"> In this study, we investigated generic summarization. An event-based scheme was employed to represent document and identify important content. The independent event-based approach identified important content according to event frequency. We also investigated the different importance of event terms in different context.</Paragraph>
    <Paragraph position="1"> Experiment showed that this idea achieved promising results. Then we explored summarization under different length limitation. We found that our independent event-based approaches acted well with longer summaries.</Paragraph>
    <Paragraph position="2"> In the relevant event-based approach, events were linked together by same or similar event terms and event elements. Experiments showed that the relevance between events can improve the performance of summarization. Compared with close related work, we achieved encouraging improvement.</Paragraph>
  </Section>
class="xml-element"></Paper>