<?xml version="1.0" standalone="yes"?>
<Paper uid="W04-3218">
  <Title>Mining Spoken Dialogue Corpora for System Evaluation and Modeling</Title>
  <Section position="3" start_page="0" end_page="0" type="metho">
    <SectionTitle>
Call-type: Request(Account Balance)
</SectionTitle>
    <Paragraph position="0"> dialog level. Clustering at the utterance is for modeling the language people use in a speci c dialog context; clustering at the dialog level allows us to characterize the whole interaction between the users and a system. In the next section we describe the corpora data structure. In section 3 we describe the clustering algorithms.</Paragraph>
    <Paragraph position="1"> In sections 4 and 5 we report on experiments and results for utterance-based and dialog clustering, respectively.</Paragraph>
  </Section>
  <Section position="4" start_page="0" end_page="0" type="metho">
    <SectionTitle>
2 Dialog corpora
</SectionTitle>
    <Paragraph position="0"> The corpora is collected from an automatic call routing system where the aim is to transfer the user to the right route in a large call center.</Paragraph>
    <Paragraph position="1"> An example dialog from a customer care application is given in Figure 1. After the greeting prompt, the speaker's response is recognized using an automatic speech recognizer (ASR).</Paragraph>
    <Paragraph position="2"> Then, the intent of the speaker is identi ed from the recognized sequence, using a spoken language understanding (SLU) component. This step can be framed as a classi cation problem, where the aim is to classify the intent of the user into one of the prede ned call-types (Gorin et al., 1997). Then, the user would be engaged in a dialog via clari cation or con rmation prompts until a nal route is determined.</Paragraph>
  </Section>
class="xml-element"></Paper>