<?xml version="1.0" standalone="yes"?>
<Paper uid="P04-1070">
  <Title>An alternative method of training probabilistic LR parsers</Title>
  <Section position="6" start_page="0" end_page="0" type="concl">
    <SectionTitle>
5 Conclusions
</SectionTitle>
    <Paragraph position="0"> We have presented a novel way of assigning probabilities to transitions of an LR automaton. Theoretical analysis and empirical data reveal the following.</Paragraph>
    <Paragraph position="1"> The efficiency of LR parsing remains unaffected. Although a right-to-left order of reading input underlies the novel training method, we may continue to apply the parser from left to right, and benefit from the favourable computational properties of LR parsing.</Paragraph>
    <Paragraph position="2"> The available space of probability distributions is significantly larger than in the case of the methods published before. In terms of the number of free parameters, the difference that we found empirically exceeds one order of magnitude. By the same criteria, we can now guarantee that LR parsers are at least as powerful as the CFGs from which they are constructed. null Despite the larger number of free parameters, no increase of sparse data problems was observed, and in fact there was a small increase in accuracy.</Paragraph>
  </Section>
class="xml-element"></Paper>