<?xml version="1.0" standalone="yes"?>
<Paper uid="C02-2017">
  <Title>Robust Interpretation of User Requests for Text Retrieval in a Multimodal Environment</Title>
  <Section position="15" start_page="2" end_page="2" type="concl">
    <SectionTitle>
6 Conclusion
</SectionTitle>
    <Paragraph position="0"> We have presented an interpretation component for natural language user input in a web-based multi-modal text retrieval system. By applying well-known and simple methods from shallow parsing and knowledge-based text retrieval and integrating them in a novel way we have succeeded in creating a robust, flexible and efficient parser for our application. null An important feature is the distinction between those parts of utterances relating to structure and those relating to content. This is achieved by taking advantage of the fact that only a limited vocabulary and set of expressions are used for the former. This allows us to employ simple rule-based techniques for their interpretation. The identification of the content on the other hand is done with the help of a co-occurrence repository, at the moment consisting of adjective-noun/proper name pairs. In the future we will have to investigate whether search results can be improved by inserting other combinations, like noun-preposition-noun triples.</Paragraph>
  </Section>
class="xml-element"></Paper>