<?xml version="1.0" standalone="yes"?>
<Paper uid="W04-3240">
  <Title>Learning to Classify Email into &amp;quot;Speech Acts&amp;quot;</Title>
  <Section position="6" start_page="112" end_page="112" type="concl">
    <SectionTitle>
5 Concluding Remarks
</SectionTitle>
    <Paragraph position="0"> We have presented an ontology of &amp;quot;email speech acts&amp;quot; that is designed to capture some important properties of a central use of email: negotiating and coordinating joint activities. Unlike previous attempts to combine speech act theory with email (Winograd, 1987; Flores and Ludlow, 1980), we propose a system which passively observes email and automatically classifies it by intention. This reduces the burden on the users of the system, and avoids sacrificing the flexibility and socially desirable aspects of informal, natural language communication.</Paragraph>
    <Paragraph position="1"> This problem also raises a number of interesting research issues. We showed that entity extraction and part of speech tagging improves classifier performance, but leave open the question of whether other types of linguistic analysis would be useful. Predicting speech acts requires context, which makes classification an inherently sequential task, and the labels assigned to messages have non-trivial structure; we also leave open the question of whether these properties can be effectively exploited.</Paragraph>
    <Paragraph position="2"> Our experiments show that many categories of messages can be detected, with high precision and moderate recall, using existing text-classification learning methods. This suggests that useful task-tracking tools could be constructed based on automatic classifiers--a potentially important practical application.</Paragraph>
  </Section>
class="xml-element"></Paper>