<?xml version="1.0" standalone="yes"?>
<Paper uid="P98-2234">
  <Title>Some Properties of Preposition and Subordinate Conjunction Attachments*</Title>
  <Section position="2" start_page="0" end_page="1436" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> Determining the attachments of prepositions and subordinate conjunctions is an important problem in parsing natural language. It is also an old problem that continues to elude a complete solution. A classic example of the problem is the sentence &amp;quot;I saw a man with a telescope&amp;quot; , where who had the telescope is ambiguous.</Paragraph>
    <Paragraph position="1"> Recently, the preposition attachment problem has been addressed using corpus-based methods (Hindle and Rooth, 1993; Ratnaparkhi * This paper reports on work performed at the MITRE Corporation under the support of the MITRE Sponsored Research Program. Useful advice was provided by Lynette Hirschman and David Palmer. The experiments made use of Morgan Pecelli's noun/verb group annotations and some of David Day's programs.</Paragraph>
    <Paragraph position="2"> et al., 1994; Brill and Resnik, 1994; Collins and Brooks, 1995; Merlo et al., 1997). The present paper follows in the path set by these authors, but extends their work in significant ways. We made these extensions to solve this problem in a way that can be directly applied in running systems in such application areas as information extraction or conversational interfaces.</Paragraph>
    <Paragraph position="3"> In particular, we have sought to produce an attachment decision procedure with far broader coverage than in earlier approaches. Most research to date has focussed on a subset of the attachment problem that only covers 25% of the problem instances in our training data, the so-called binary VNP subset. Even the broader V\[NP\]* subset addressed by (Merlo et al., 1997) only accounts for 33% of the problem instances.</Paragraph>
    <Paragraph position="4"> In contrast, our approach attempts to form attachments for as much as 89% of the problem instances (modulo some cases that are either pathological or accounted for by other means).</Paragraph>
    <Paragraph position="5"> Work to date has also been concerned primarily with reproducing the structure of Tree-bank annotations. In other words, the underlying syntactic paradigm has been the traditional notion of full sentential parsing. This approach differs from the parsing models currently being explored by both theorists and practitioners, which include semi-parsing strategies and finite-state approximations to context-free grammars. Our approach to syntax uses a cascade of rule sequence processors, each of which can be thought of as approximating some aspect of the underlying grammar by finite-state transduction. We have thus had to extend previous work at the conceptual level as well, by recasting the preposition attachment problem in terms of the vocabulary of finite-state approximations (noun groups, etc.), rather than the traditional syntactic categories (noun phrases, etc.).</Paragraph>
    <Paragraph position="6">  Much of the present paper is thus concerned with describing our extensions to the preposition attachment problem. We present the problem scope of interest to us, as well as the data annotations required to support our investigation. We also present a decision procedure for attaching prepositions and subordinate conjunctions. The procedure is trained through error-driven transformation learning (Brill, 1993), and we present a number of training experiments and report on the performance of the trained procedure. In brief, on the restricted VNP problem, our procedure achieves nearly the same level of test-set performance (83.1%) as current state-of-the-art systems (84.5% (Collins and Brooks, 1995)).</Paragraph>
    <Paragraph position="7"> On the unrestricted data set, our procedure achieves an attachment accuracy of 75.4%.</Paragraph>
  </Section>
class="xml-element"></Paper>