<?xml version="1.0" standalone="yes"?>
<Paper uid="P95-1054">
  <Title>Quantifying lexical influence: Giving direction to context</Title>
  <Section position="7" start_page="332" end_page="332" type="evalu">
    <SectionTitle>
4 Discussion
</SectionTitle>
    <Paragraph position="0"> Of the several different types of word-level associations, lexical and lexico-semantic associations are among the most significant local associations. Lexical (or associative) context is characterised by rigid word order, and usually implies that the primer and the primed together act as one lexical unit. Lexico-semantic associations are exemplified by phrasal verbs (eg, 'fix up'), and are characterised by morphological complexity in the verb part and spatial flexibility in the phrase as a whole.</Paragraph>
    <Paragraph position="1"> It is noteworthy that all the three DIM's capture the notions of lexical (ie, fixed) and lexico-semantic associations in one formula (albeit to differing degrees of success). Thus we have 'staff' and 'reporter' influencing each other almost equally, while the asymmetric influence on 'in' from its right context ('addition') is also detected by the DIM's.</Paragraph>
    <Paragraph position="2"> It is our contention that symmetric measures constrain the re-ranking/proposing process significantly, since they are essentially blind to a significant fraction (perhaps more than ha/f) of all co-occurrence phenomena in natural language.</Paragraph>
  </Section>
class="xml-element"></Paper>