<?xml version="1.0" standalone="yes"?>
<Paper uid="P94-1014">
  <Title>AN EMPIRICAL MODEL OF ACKNOWLEDGMENT FOR SPOKEN-LANGUAGE SYSTEMS</Title>
  <Section position="9" start_page="99" end_page="100" type="concl">
    <SectionTitle>
4 CONCLUSION
</SectionTitle>
    <Paragraph position="0"> Why is a conversation-analytic study of acknowledgment useful in the development of spoken language systems? SLSs developers face the dual challenges of creating both domain-based dialogues and repair-oriented dialogues. Lacking systematic mechanisms for natural maintenance of mutuality, SLSs tend to rely on domain structures--producing rather stolid interaction. The most advanced systems incorporate repair acts, but are unable to relate the repairs to the main dialogue structures in a natural way. The acknowledgment model described in this paper provides a systematic method of maintaining mutuality of knowledge for both domain and control information.</Paragraph>
    <Paragraph position="1"> More concretely, using this model SLSs can account for acknowledgments by both user and system. The corpus evidence suggests that users' utterances in unconstrained dialogues contain many instances of acknowledgment. In interpreting these utterances, identification of the appropriate acknowledgment function affects the state of the dialogue model and thus plays an important role in determining an appropriate response by the system. In producing such responses, the acknowledgment model can provide structurally appropriate utterances. The fundamental idea is to produce contextually appropriate acknowledgments that advances the dialogue seamlessly with respect to both domain and control functions. That is, the system needs to give the right signals at the right time.</Paragraph>
    <Paragraph position="2"> The evidence of the U S WEST VNS corpus suggests that understanding and production of domain and control utterances are closely linked; they thus cannot be implemented as independent mechanisms in an SLS. For example, giving directions involves presenting large amounts of information for which an installment approach often proved effective. Typically the user was given the opportunity to choose the style of presentation of  directions, either step-by-step or all at once. The choice of presentation method by the conversants was a dynamic one: in cases where it became apparent that the user was experiencing difficulties with either hearing or understanding directions, the wizard often resorted to the step-by-step approach. This form of repair changed the process of interaction so that the comprehension of each installment was verified before proceeding with the next.</Paragraph>
    <Paragraph position="3"> The conversants in the VNS corpus displayed relatively higher rates of use of acknowledgment in repair situations or when unplanned events arose (e.g., the user had gotten lost). Intuitively, people make more effort to establish mutual knowledge when it is apparent that miscommunication has occurred than at other times; their certainty criterion for mutuality (Clark and Marshall, 1981) is raised as a result of the need for repair. This suggests that a facility for acknowledgment is an important element in the development of robust SLSs because use of acknowledgment is most critical precisely when the conversation has gone awry. We are currently developing a computational model of acknowledgment based on the empirical work presented in this paper. This model is intended for integration into a SLS where it will serve both to predict when acknowledgments are appropriate from the system and when to expect acknowledgments from the user. Briefly, determining the applicability of an acknowledgment involves interpreting exchanges in terms of speech acts and then mapping these speech-act patterns onto the acknowledgment classes described. This, we believe, will facilitate improved SLS robustness through achievement of a greater degree of mutual understanding and provide a more natural and intuitive interaction. The utility and implementation of the empirical model will be the focus of a later paper.</Paragraph>
  </Section>
class="xml-element"></Paper>