<?xml version="1.0" standalone="yes"?>
<Paper uid="H01-1021">
  <Title>Evaluating Question-Answering Techniques in Chinese</Title>
  <Section position="8" start_page="0" end_page="0" type="concl">
    <SectionTitle>
6. Summary and Future Work
</SectionTitle>
    <Paragraph position="0"> The evaluations on Marsha, although preliminary, indicate that techniques developed for question answering in English are also effective in Chinese. In future research, we plan to continue to improve these techniques and carry out more careful evaluations to establish whether there are any significant differences in the question-answering task between these two languages.</Paragraph>
    <Paragraph position="1"> The evaluation of the English version of Marsha indicates that the Marsha heuristics work well in English as well as in Chinese. We now plan to incorporate these techniques in a cross-lingual question-answering system for English and Chinese. By using two systems with similar question processing strategies, we hope to exploit the query templates to produce accurate question translations.</Paragraph>
    <Paragraph position="2"> We have also started to develop a probabilistic model of question answering using the language model approach (Ponte and Croft, 1998). This type of model will be essential for extending the capability of QA systems beyond a few common query forms.</Paragraph>
  </Section>
class="xml-element"></Paper>