<?xml version="1.0" standalone="yes"?>
<Paper uid="A88-1014">
  <Title>BUILDING A LARGE THESAURUS FOR INFORMATION RETRIEVAL</Title>
  <Section position="5" start_page="106" end_page="106" type="concl">
    <SectionTitle>
6. Conclusions and Future Work
</SectionTitle>
    <Paragraph position="0"> Based on preliminary investigations, it appears that lexical-semantic relations can be used to give small improvements in the effectiveness of information retrieval systems. Work with Webster's Seventh New Collegiate Dictionary and the Collins Dictionary of the English Language has demonstrated that text processing and natural language parsing techniques can be used to extract and organize important data about English that will be of value for information retrieval and for a variety of natural language processing applications. Furthermore, it is clear that the SNePS semantic network system can be used to store that type of data in a form that will permit both rule and path-based inference.</Paragraph>
    <Paragraph position="1"> Future work will include further processing of dictionaries (including the Longman Dictionary of Contemporary English and others from the Oxford Text Archive), merging the resulting output into a large semantic network, extending the capabilities of SNePS to handle a very large thesaurus, integrating SNePS with the SMART and CODER systems, and further testing of the utility of that thesaurus to support interactive information retrieval sessions.</Paragraph>
  </Section>
class="xml-element"></Paper>