<?xml version="1.0" standalone="yes"?>
<Paper uid="A94-1028">
  <Title>Robust Text Processing in Automated Information Retrieval</Title>
  <Section position="7" start_page="171" end_page="172" type="concl">
    <SectionTitle>
6 Conclusions
</SectionTitle>
    <Paragraph position="0"> We presented in some detail our natural language information retrieval system consisting of an advanced NLP module and a 'pure' statistical core engine. While many problems remain to be resolved, including the question of adequacy of term-based representation of document content, we attempted to demonstrate that the architecture described here is nonetheless viable. In particular, we demonstrated that natural language  with 1000 does retrieved per query: (1) txtl - single terms of &lt;narr&gt; and &lt;desc&gt; fields m this is the base ran; (2) txt2 - &lt;hart&gt; and &lt;desc&gt; fields with low weight terms removed; (3) txt2+nlp -&lt;narr&gt; and &lt;desc&gt; fields including syntactic phrase terms using the new weighting scheme; (4) con - &lt;desc&gt; and &lt;con&gt; fields with low weight terms removed but with no NLP; and (5) con+nip - &lt;dese&gt; and &lt;con&gt; fields including phrases with the new weighting scheme.</Paragraph>
    <Paragraph position="1"> processing can now be done on a fairly large scale and that its speed and robustness can match those of traditional statistical programs such as key-word indexing or statistical phrase extraction. We suggest, with some caution until more experiments are run, that natural language processing can be very effective in creating appropriate search queries out of a user's initial specifications, which can be frequently imprecise or vague.</Paragraph>
  </Section>
class="xml-element"></Paper>