<?xml version="1.0" standalone="yes"?>
<Paper uid="W02-1005">
  <Title>Augmented Mixture Models for Lexical Disambiguation</Title>
  <Section position="2" start_page="0" end_page="0" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> The focus tasks of this paper are two related problems in lexical ambiguity resolution:  Word Sense Disambiguation (WSD) and Context-Sensitive Spelling Correction (CSSC). Word Sense Disambiguation has a long history as  a computational task (Kelly and Stone, 1975), and the field has recently supported large-scale international system evaluation exercises in multiple languages (SENSEVAL-1, Kilgarriff and Palmer (2000), and SENSEVAL-2, Edmonds and Cotton (2001)).</Paragraph>
    <Paragraph position="1"> General purpose Spelling Correction is also a long-standing task (e.g. McIlroy, 1982), traditionally focusing on resolving typographical errors such as transposition and deletion to find the closest &amp;quot;valid&amp;quot; word (in a dictionary or a morphological variant), typically ignoring context. Yet Kukich (1992) observed that about 25-50% of the spelling errors found in modern documents are either context-inappropriate misuses or substitutions of valid words (such as principal and principle) which are not detected by traditional spelling correctors. Previous work has addressed the problem of CSSC from a machine learning perspective, including Bayesian and Decision List models (Golding, 1995), Winnow (Golding and Roth, 1996) and Transformation-Based Learning (Mangu and Brill, 1997).</Paragraph>
    <Paragraph position="2"> Generally, both tasks involve the selection between a relatively small set of alternatives per key-word (e.g. sense id's such as church/BUILDING and church/INSTITUTION or commonly confused spellings such as quiet and quite), and are dependent on local and long-distance collocational and syntactic patterns to resolve between the set of alternatives. Thus both tasks can share a common feature space, data representation and algorithm infrastructure. We present a framework of doing so, while investigating the use of mixture models in conjunction with a new error-correction technique as competitive alternatives to Bayesian models. While several authors have observed the fundamental similarities between CSSC and WSD (e.g. Berleant, 1995 and Roth, 1998), to our knowledge no previous comparative empirical study has tackled these two problems in a single unified framework.</Paragraph>
  </Section>
class="xml-element"></Paper>