<?xml version="1.0" standalone="yes"?>
<Paper uid="P96-1003">
  <Title>Noun-Phrase Analysis in Unrestricted Text for Information Retrieval</Title>
  <Section position="10" start_page="22" end_page="23" type="evalu">
    <SectionTitle>
5 Results
</SectionTitle>
    <Paragraph position="0"> The results of the experiment are given in Tables 1, 2, and 3. In general, we see improvement in both recall and precision.</Paragraph>
    <Paragraph position="1"> Recall improves slightly (about 1%), as shown in Table 1. While the actual improvement is not significant for the run of fifty queries, the increase in absolute numbers of relevant documents returned indicates that the small compounds supported better matches in some cases.</Paragraph>
    <Paragraph position="2"> Interpolated precision improves significantly5 as shown in Table 2. The general improvement in precision indicates that small compounds provide more accurate (and effective) indexing terms than full NPs.</Paragraph>
    <Paragraph position="3"> Precision improves at various returned-document levels, as well, as shown in Table 3. Initial precision, in particular, improves significantly. This 1degNote that the CLARIT process used as a baseline does not reflect optimum CLARIT performance, e.g., as obtained in actual TREC evaluations, since we did not use a variety of standard CLARIT techniques that significantly improve performance, such as automatic query expansion, distractor space generation, subterm indexing, or differential query-term weighting. Cf. (Evans et al., 1996) for details.</Paragraph>
    <Paragraph position="4">  The PES, which was not optimized for processing, required approximately 3.5 hours per 20megabyte subset of AP89 on a 133-MHz DEC alpha processor) 3 Most processing time (more than 2 of every 3.5 hours) was spent on simplex NP parsing.</Paragraph>
    <Paragraph position="5"> Such speed might be acceptable in some, smaller-scale IR applications, but it is considerably slower than the baseline speed of CLARIT noun-phrase identification (viz., 200 megabytes per hour on a 100-MIPS processor).</Paragraph>
    <Paragraph position="6"> l~ (Evans et al., 1995; Evans et al., 1996) 13Note that the machine was not dedicated to the PES processing; other processes were running simultaneously.</Paragraph>
  </Section>
class="xml-element"></Paper>