<?xml version="1.0" standalone="yes"?>
<Paper uid="W00-1310">
  <Title>Nonlocal Language Modeling based on Context Co-occurrence Vectors</Title>
  <Section position="9" start_page="85" end_page="85" type="concl">
    <SectionTitle>
6 Conclusion
</SectionTitle>
    <Paragraph position="0"> In this paper we described a novel language model of incorporating long-distance lexical dependencies based on context co-occurrence vectors. Reduced vector representation of word co-occurrences enables rather simple but effective representation of the context. Significant reductions in perplexity are obtained relative to a staaldard trigram model: both on the entire text. (5.0~) and on the target vocabulary (27.2%).</Paragraph>
  </Section>
class="xml-element"></Paper>