<?xml version="1.0" standalone="yes"?>
<Paper uid="P91-1012">
  <Title>COMPOSE-REWUCE PARSING</Title>
  <Section position="10" start_page="95" end_page="96" type="concl">
    <SectionTitle>
IV. CONCLUSIONS
</SectionTitle>
    <Paragraph position="0"> Aside from the intrinsic interest in the abstract of real-time parsablility, is there any practical significance to these results. Two drawbacks, one already referred to, certainly restrict their significance. One is that the restriction to atomic category CF-PSGs is crucial the fact that the comparison between a rule element and a node label is atomic and constant time is fundamental. Any move to features or other annotations would put an end to real-time processing. This fact gives added weight to the problem mentioned above in section II,4, that only left-common analysis results are shared between alternatives. Thus if one finesses the atomic category problem by using a parser such as those described here only as the first pass of a two pass system, one is only putting off the payment of the complexity price to the second pass, in the absence to date of any linear time solution to the constraint satisfaction problem. On this basis, one would clearly prefer a parallel CKY/Earley algorithm, which does share all common substructure, to the parsers presented here.</Paragraph>
    <Paragraph position="1"> Nevertheless, there is one class of applications where the left-to-right real-time behaviour of these algorithms may be of practical benefit, namely in speech recognition.</Paragraph>
    <Paragraph position="2"> Present day systems require on-line availability of syntactic and domainsemantic constraint to limit the search space at lower levels of the system. Hitherto this has meant these constraints must be brought to bear during recognition as some form of regular grammar, either explicitly  constructed as such or compiled into.</Paragraph>
    <Paragraph position="3"> The parsers presented here offer the alternative of parallel application of genuinely context-free grammars directly, with the potential added benefit that, with sufficient processor width, quite high degrees of local ambiguity can be tolerated, such as would arise if (a finite subset of) a feature-based grammar were expanded out into atomic category form.</Paragraph>
  </Section>
class="xml-element"></Paper>