<?xml version="1.0" standalone="yes"?>
<Paper uid="W02-1014">
  <Title>Fast LR Parsing Using Rich (Tree Adjoining) Grammars</Title>
  <Section position="5" start_page="0" end_page="0" type="concl">
    <SectionTitle>
5 Conclusions
</SectionTitle>
    <Paragraph position="0"> The results presented here suggest that: (1) the use of a rich grammar as the underlying formalism for the LR techniques makes available enough information to the driver so as to allow for a greedy strategy to achieve reasonable parsing accuracy. (2) LR parsing allows for very fast parsing with at least reasonable accuracy.</Paragraph>
    <Paragraph position="1"> The approach seems to have much yet to be explored, mostly to improve the accuracy side. In particular we have not yet come with a solid approach to lexicalization. Using words (as opposed to pos tags) as the terminals of the grammar to be pre-compiled leads to an explosion in the size of the table: not only the average number of transitions per state grows, but also the number of states itself grows wildly. One very promising approach for a partial solution is to expand the set of terminals by adding some selected syntactic sub-categories that have distinguished syntactic behavior, as we reported in this paper for time nouns, or by individuating frequent words with peculiar behavior, as we did for the word &amp;quot;that&amp;quot;. Although we have also done some initial work on a more general approach to clustering words according to their syntactic distribution, they are not still adequate for our purposes. Finally, an earlier simple experiment of adding a dependency on the lookahead's word (recall that a23a25a24 in  a19 was the pos tag only), gave us a small improvement of about a couple of percents in the accuracy measures.</Paragraph>
    <Paragraph position="2"> A limited amount of parallelism is an important topic to be considered, perhaps together with a better notion (non-binary) of confidence. The high reliability of a27a12a24a31a29 a0 a19 , suggests that we should look for a way to enrich the parsing table.</Paragraph>
    <Paragraph position="3"> LR parser for the full class of TAGs is problematic. The bpack action of early structural commitment is involved in most of the decision points where the wrong action is taken. We are currently working on a version of the LR parser for a subclass of TAGs, the Tree Insertion Grammars (Schabes and Waters, 1995), for which efficient true LR parsers can be obtained.</Paragraph>
  </Section>
class="xml-element"></Paper>