<?xml version="1.0" standalone="yes"?>
<Paper uid="P03-2028">
  <Title>Spoken Interactive ODQA System: SPIQA</Title>
  <Section position="6" start_page="0" end_page="0" type="concl">
    <SectionTitle>
4 Conclusion
</SectionTitle>
    <Paragraph position="0"> The proposed spoken interactive ODQA system, SPIQA copes with missing information by adopting disambiguation of users' questions by system queries. In addition, a speech summarization technique was applied for handling recognition errors.</Paragraph>
    <Paragraph position="1"> Although adding information was performed using DQs only once, experimental results revealed the potential of the generated DQs to acquire indispensable information that was lacking for extracting answers. In addition, the screening filter helped to generate the appropriate DQs. Future research will in- null An integer without a % other than MRRs indicates number of sentences. Word acc.:word accuracy, SPK:speaker, AVG: averaged values, w/o errors: transcribed sentences without recognition errors, APP: appropriate DQs and InAPP: inappropriate DQs.</Paragraph>
    <Paragraph position="2"> clude an evaluation of the appropriateness of DQs derived repeatedly to obtain the final answers. In addition, the interaction strategy automatically generated by the DDQ module should be evaluated in terms of how much the DQs improve QA's total performance. null</Paragraph>
  </Section>
class="xml-element"></Paper>