<?xml version="1.0" standalone="yes"?>
<Paper uid="W01-0719">
  <Title>Combining Linguistic and Machine Learning Techniques for Email Summarization</Title>
  <Section position="10" start_page="0" end_page="0" type="concl">
    <SectionTitle>
6 Conclusions and Future Work
</SectionTitle>
    <Paragraph position="0"> In this paper we presented a novel technique for document gisting suitable for domain and genre independent collections such as email messages.</Paragraph>
    <Paragraph position="1"> The method extracts simple noun phrases using linguistictechniques and then uses machine learning to classify them as salient for the document content. The contributions of this work are: 1. From a linguistic standpoint, we demonstrated that the modifiers of a noun phrase can be as semantically important as the head for the task of gisting.</Paragraph>
    <Paragraph position="2"> 2. From a machine learning standpoint, we evaluated the power and limitation of several classifiers: decision trees, rule induction, and decision forests classifiers.</Paragraph>
    <Paragraph position="3"> 3. We proved that linguistic knowledge can enhance machine learning by evaluating the impact of linguistic filtering before applying the learning scheme.</Paragraph>
    <Paragraph position="4"> The study, the evaluation, and the results provide experimental grounds for research not only in summarization, but also in information extraction and topic detection.</Paragraph>
  </Section>
class="xml-element"></Paper>