<?xml version="1.0" standalone="yes"?>
<Paper uid="W02-0905">
  <Title>Using Co-Composition for Acquiring Syntactic and Semantic Subcategorisation</Title>
  <Section position="2" start_page="0" end_page="0" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> Recent lexicalist Grammars project the subcategorisation information encoding in the lexicon onto syntactic structures. These grammars use accurate subcategorised lexicons to restrict potential syntactic structures. In terms of parsing development, it is broadly assumed that parsers need such information in order to reduce the number of possible analyses and, therefore, solve syntactic ambiguity. Over the last years various methods for acquiring subcategorisation information from corpora has been proposed. Some of them induce syntactic subcategorisation from tagged texts (Brent, 1993; Briscoe and Carrol, 1997; Marques, 2000). Unfortunately, syntactic information is not enough to solve structural ambiguity. Consider the following verbal phrases: (1) [peel [a2a4a3 the potato] [a3a5a3 with a knife]] (2) [peel [a2a4a3 [a2a6a3 the potato] [a3a7a3 with a rough stain]]] The attachment of &amp;quot;with PP&amp;quot; to both the verb &amp;quot;peel&amp;quot; in phrase (1) and to the NP &amp;quot;the potato&amp;quot; in (2) does not depend only on syntactic requirements.</Paragraph>
    <Paragraph position="1"> Indeed, it is not possible to attach the PP &amp;quot;with a knife&amp;quot; to the verb &amp;quot;peel&amp;quot; by asserting that this verb subcategorises a &amp;quot;with PP'. Such a subcategorisation information cannot be used to explain the analysis of phrase (2), where it is the NP &amp;quot;the potato&amp;quot; that is attached to the &amp;quot;with PP&amp;quot;. In order to decide the correct analysis in both phrases, we are helped by our world knowledge about the action of peeling, the use of knifes, and the attributes of potatoes. In general, we know that knifes are used for peeling, and potatoes can have different kinds of stains. So, the parser is able to propose a correct analysis only if the lexicon is provided with, not only syntactic subcategorisation information, but also with information on semantic-pragmatic requirements (i.e., with selection restrictions).</Paragraph>
    <Paragraph position="2"> Other works attempt to acquire selection restrictions requiring pre-existing lexical ressources. The learning algorithm requires sample corpora to be constituted by verb-noun, noun-verb, or verb-prepnoun dependencies, where the nouns are semantically tagged by using lexical hierarchies such as WordNet (Resnik, 1997; Framis, 1995). Selection restrictions are induced by considering those dependencies associated with the same semantic tags. For instance, if verb ratify frequently appears with nouns semantically tagged as &amp;quot;legal documents&amp;quot; in the direct object position (e.g., article, law, precept, . . . ), then it follows that it must select for nouns denoting legal documents. Unfortunately, if a pre-defined July 2002, pp. 34-41. Association for Computational Linguistics.</Paragraph>
  </Section>
class="xml-element"></Paper>