<?xml version="1.0" standalone="yes"?>
<Paper uid="W06-1909">
  <Title>Adapting a Semantic Question Answering System to the Web</Title>
  <Section position="6" start_page="67" end_page="67" type="concl">
    <SectionTitle>
6 Conclusion and Perspectives
</SectionTitle>
    <Paragraph position="0"> In this paper, an existent textual QA system was extended and modified to work successfully on the German web as a virtual document collection.</Paragraph>
    <Paragraph position="1"> The main results are: precision-oriented extensions and experimentally derived parameter settings are needed to achieve similar performance on the vast web as on small-sized document collections that show higher homogeneity and quality of the contained texts; taking a semantic QA system to the web is feasible as demonstrated in this paper, but answering a question is still expensive in terms of bandwidth and CPU time.</Paragraph>
    <Paragraph position="2"> There are several interesting directions for futurework. ThefirstlevelofInSicht-W3canbeimproved by finding a better suited search engine or by building and running a new one in a distributed manner. Ideally, it should support arbitrarily complex Boolean queries, parameterized proximity operators like NEAR/N (N [?] {1,2,...,100}), or even linguistically informed operators like same-sentence and same-paragraph.</Paragraph>
    <Paragraph position="3"> The second level (the textual QA system) can be improved by acquiring more inferential knowledge to allow better query expansion. The matching approach can be extended from the unit sentencetoalargerlinguisticunitlikeparagraph, text, and even text collection. Distributed architectures and algorithms can reduce answer times.</Paragraph>
  </Section>
class="xml-element"></Paper>