<?xml version="1.0" standalone="yes"?>
<Paper uid="W00-1015">
  <Title>Flexible Speech Act Based Dialogue Management</Title>
  <Section position="8" start_page="138" end_page="138" type="concl">
    <SectionTitle>
6 Conclusion
</SectionTitle>
    <Paragraph position="0"> We have presented some results from our research on spoken dialogue management. We concentrated on how to dynamically calculate a collection of predictions for how to continue a dialogue (dialogue primitives), how to account for different dialogue strategies and utterances with several communicative goals through combinations of primitives, and how to map the user utterances onto primitives. The approach has been implemented and tested in several prototype systems, e.g., horoscope, movie, and telephone rate service (Feldes et al., 1998).</Paragraph>
    <Paragraph position="1"> Dialogue grammars have previously been used to manage dialogue (Bunt, 1989; Bilange, 1991; Traum and Hinkelman, 1992; JSnsson, 1993; Mast et al., 1994; Novick and Sutton, 1994; Chino and Tsuboi, 1996), but we are not aware of an approach where speech acts are translated into a collection of primitives with propositional content.</Paragraph>
    <Paragraph position="2"> Previous grammar approaches use the speech acts directly or assume a one-to-one correspondence between utterance and speech act.</Paragraph>
    <Paragraph position="3"> Through the natural division of the knowledge into type and content, we have achieved a flexible dialogue manager that adapts to users' behaviour. We can take advantage of the predictive capabilites of speech act grammars and still be able to account for multi-functional utterances.</Paragraph>
    <Paragraph position="4"> We have also demonstrated that our approach is flexible: 1. the dialogue engine, the pragmatic interpreter, the primitives and the algorithm for mapping user utterances onto predictions are application and language independent, which makes it easy to reuse our dialogue manager in new applications, and 2. the dialogue manager can easily account for several types of dialogue, e.g., strict question-answer or mixed initiative. We give the service designer the freedom to decide which kind of dialogue she wants---on a high level--and the dialogue manager combines the basic primitives accordingly.</Paragraph>
    <Paragraph position="5"> Future work includes empirical testing to verify whether we are calculating appropriate predictions. Also, several aspects of our dialogue grammar have not yet been translated into primitives, for example, the frequent use of assert in natural dialogue. As a wider dialogue coverage is required, we will add primitives accordingly. We are also working on using the primities as input to a multi-lingual automatic text generation system.</Paragraph>
  </Section>
class="xml-element"></Paper>