<?xml version="1.0" standalone="yes"?>
<Paper uid="C96-2190">
  <Title>Prepositional Phrase Attachment Through A Hybrid Disambiguation Model</Title>
  <Section position="3" start_page="0" end_page="0" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> The resolution of prepositional phrase attachment ambiguity is a difficult problem in NLP. There have been many proposals to attack this problem. Traditional proposMs are mainly based on knowledge-based techniques which heavily depend on empirical knowledge encoded in handcrafted rules and domain knowledge in knowledge base: they are therefore not scalable. Recent work has turned to corpus-based or statistical approaches (e.g. Hindle and Rooth 1993; Ratnaparkhi, Reynar and Roukos 1994, Brill and Resnik 1994, Collins and Brooks 1995). Unlike traditional proposals, corpus-based approaches need not to prepare a large amount of handcrafted rules, they have therefore the merit of being scalable or easy to transfer to new domains. However, corpus-based approaches shffer fi'om the notorious sparse-data problem: estimations based on low occurrenee frequencies are very unreliable and often result in bad performances in disambiguation. To cope with this problem, Brill and Resnik (1994) use word classes from Word-Net noun hierarchy to (:luster words into semantic classes. Collins and Brooks (1995) on the other hand use morphological analysis t)oth on test and tr~fining data. Unfortunately, all these smoothing methods are not efficient enough to make a significant improvement on perforlnancc.</Paragraph>
    <Paragraph position="1"> Instead of using pure statistical approaches stated above, wc propose a hybrid approach to attack PP attachment problem. We employ corpus-based likelihood analysis to choose most-likely attachment. Where the occurrence frequency is too low to make a reliable choice, wc turn to use conceptual infornlation froln a machine-readable dictionary to to make decision on PP attachments.</Paragraph>
    <Paragraph position="2"> We use this disambiguation method to buihl a disambiguation module in PFTE system, l In what follows we first outline the idea of using hybrid information to sui)ply preferences for resolving ambiguous PP attachment. We then describe how this information is used in disambiguating PP attachment. We put the hybrid approach in an disambiguation algorithm. Finally, we show an experiment and its result.</Paragraph>
  </Section>
class="xml-element"></Paper>