<?xml version="1.0" standalone="yes"?>
<Paper uid="H01-1040">
  <Title>Intelligent Access to Text: Integrating Information Extraction Technology into Text Browsers</Title>
  <Section position="7" start_page="0" end_page="0" type="concl">
    <SectionTitle>
5. CONCLUSIONS
</SectionTitle>
    <Paragraph position="0"> To date IE has largely been a &amp;quot;technology push&amp;quot; activity, with language engineers working to develop core technologies. For the technology to become usable, and for its further development to be influenced by end user requirements (&amp;quot;user pull&amp;quot;), prototype end-user application systems must be built which exploit the significant achievement of the technology to date, while acknowledging its limitations. In this paper we have described such a prototype, the TRESTLE system, which exploits named entity and scenario template IE technology to offer users novel ways to access textual information.</Paragraph>
    <Paragraph position="1"> Our preliminary evaluation has revealed that although search options initially selected from the access frame were not always optimal for undertaking set tasks, the colour coded textual and iconic cues embedded in the headline index and full text frames on the whole enabled users to exploit the different functions seamlessly.</Paragraph>
    <Paragraph position="2"> Whilst the TRESTLE interface appeared to support interaction at a procedural level, at the conceptual level however, searchers did not necessarily gain sufficient understanding of the underlying functionality, particularly in respect to the scenario access. For example the inability to distinguish between the original headlines and the system generated summaries for SCAT was problematic and requires further investigation. Other studies have reported similar issues in introducing more complex interactive search functions [3, 8]. More meaningful labelling may in part address some of the difficulties encountered. A more extensive evaluation in a work setting will follow to assess to what extent the integration of new and established conventions can support users with domain knowledge and greater familiarity with alerting systems to adopt new searching and awareness approaches effectively.</Paragraph>
  </Section>
class="xml-element"></Paper>