<?xml version="1.0" standalone="yes"?>
<Paper uid="C02-1081">
  <Title>Data-driven Classification of Linguistic Styles in Spoken Dialogues</Title>
  <Section position="8" start_page="0" end_page="0" type="concl">
    <SectionTitle>
6 Conclusion
</SectionTitle>
    <Paragraph position="0"> This investigation showed that a4 numerical parameter values can be computed for speakers in spoken dialogue corpora, a4 these parameters can be reduced to linguistically interpretable factors by means of a PCA, a4 stable classes can be constructed from these factors by cluster analysis, a4 unseen class members can be reliably classified by trained neural networks if the data is linguistically rich, a4 style-specific language models reduce perplexity only marginally.</Paragraph>
    <Paragraph position="1"> This process has been applied to three different corpora. It has been shown that it works in principle. Further improvements may be obtained by optimizing the procedure according to specific needs (e.g. very quick classification, recognizing a speaker from a small set of possible speakers) which depend on the application.</Paragraph>
    <Paragraph position="2"> The methods can not only be applied to classify speakers according to their style, but also to recognize text genre or speech act types.</Paragraph>
  </Section>
class="xml-element"></Paper>