<?xml version="1.0" standalone="yes"?>
<Paper uid="W00-1016">
  <Title>Dialogue Helpsystem based on Flexible Matching of User Query with Natural Language Knowledge Base</Title>
  <Section position="9" start_page="148" end_page="148" type="concl">
    <SectionTitle>
8 Conclusion
</SectionTitle>
    <Paragraph position="0"> This paper described the dialogue helpsystern, which has been working in practice with real users.</Paragraph>
    <Paragraph position="1"> Construction of natural language knowledge base needs some cost, though it is much easier than that of formal language knowledge base. However, providing a high-quality service needs cost; good manuals and FAQs are important for any products, and a large amount of materials are prepared for customer service operators. With that in mind, we can say preparing a good document is a universal problem, not just to a dialogue system. null By rlmning the system, the real dialogue database can be accumulated. Based on this database, we would like to study the phenomena of man-machine conversation and to extend our work to user modeling, user intention estimation, and other interesting dialogue research areas.</Paragraph>
    <Paragraph position="2"> The system is designed to be domain-independent and can be ported to a new domain by preparing a domain knowledge base.</Paragraph>
    <Paragraph position="3"> Exploiting this merit, we are planning to construct the automatic reference service system of Kyoto University Library, which certainly provides us with a wider breadth of dialogue data.</Paragraph>
  </Section>
class="xml-element"></Paper>