<?xml version="1.0" standalone="yes"?>
<Paper uid="C90-3062">
  <Title>Repair Work in Human-Computer Dialogue</Title>
  <Section position="2" start_page="0" end_page="0" type="abstr">
    <SectionTitle>
1. Introduction
</SectionTitle>
    <Paragraph position="0"> The importance of repair in human interaction is increasingly recognised. If a dialogue is to proceed smoothly it is vital that there are opportanities for checking understanding and providing clarification when misunderstanding does occur. Everyday interaction is full of such checks and repairs, though these may be so au~;omatic as to be almost transparent, rarely disturbing ~he flow of the interaction.</Paragraph>
    <Paragraph position="1"> In human-computer interaction, providing opportunities for clarification may be even more important. If the communication is to be robust and effective, then there must be opportunities for both parties to 'repair' the interaction when it fails. If the user is communicatdeg ing in natural language (or even in a complex command language) then there are many cases where the system may not 'understand'. If the system is giving complex instructions or explanations, then there are many cases where the user may not understand.</Paragraph>
    <Paragraph position="2"> These checks and repairs have been studied by people working in the field of conversation analysis (CA) for many years. For example, people have analysed preferences for different types of repair \[7\], and typical sequences of repair moves.</Paragraph>
    <Paragraph position="3"> Recently, there has been some interest in checks and repafi&amp;quot; within Cognitive Science (though the approach *Supported by a post-doctoral fellowship from the Science and</Paragraph>
    <Section position="1" start_page="0" end_page="0" type="sub_section">
      <SectionTitle>
Engineering Research Council
</SectionTitle>
      <Paragraph position="0"> to the subject is often very different from that of CA).</Paragraph>
      <Paragraph position="1"> This includes work by Ringle and Bruce \[5\], who analyse checking moves and conversation failure, and Clark and Schaefer \[2\], who have recently proposed a model of dialogae based on contributions rather than single communicative acts. These are the sections of discourse through which the participants arrive at the mutual knowledge that the conveyed message is understood, and may involve checking and repair work.</Paragraph>
      <Paragraph position="2"> Despite the prevalence of checks and repairs in human interaction, there has been very little work within computational linguistics on these essential components of conversation. The rest of this paper will discuss the problem in more detail, and present some examples of different types of repair work in an implemented interactive explanation system.</Paragraph>
    </Section>
  </Section>
class="xml-element"></Paper>