<?xml version="1.0" standalone="yes"?>
<Paper uid="P06-3010">
  <Title>Sydney, July 2006. c(c)2006 Association for Computational Linguistics A Hybrid Relational Approach for WSD - First Results</Title>
  <Section position="3" start_page="0" end_page="0" type="intro">
    <SectionTitle>
1 Introduction
Word Sense Disambiguation (WSD) is concerned
</SectionTitle>
    <Paragraph position="0"> with the identification of the correct sense of an ambiguous word given its context. Although it can be thought of as an independent task, its importance is more easily realized when it is applied to particular tasks, such as Information Retrieval or Machine Translation (MT). In MT, the application we are focusing on, a WSD (or translation disambiguation) module should identify the correct translation for a source word when options with different meanings are available.</Paragraph>
    <Paragraph position="1"> As shown by Vickrey et al. (2005), we believe that a WSD module can significantly improve the performance of MT systems, provided that such module is developed following specific requirements of MT, e.g., employing multilingual sense repositories. Differences between monolingual and multilingual WSD are very significant for MT, since it is concerned only with the ambiguities that appear in the translation (Hutchins and Sommers, 1992).</Paragraph>
    <Paragraph position="2"> In this paper we present a novel approach for WSD, designed focusing on MT. It follows a hybrid strategy, i.e., knowledge and corpus-based, and employs a highly expressive relational formalism to represent both the examples and background knowledge. This approach allows the exploitation of several knowledge sources, together with evidences provided by examples of disambiguation, both automatically extracted from lexical resources and sense tagged corpora.</Paragraph>
    <Paragraph position="3"> This is achieved using Inductive Logic Programming (Muggleton, 1991), which has not been exploited for WSD so far. In this paper we investigate the disambiguation of 7 highly ambiguous verbs in English-Portuguese MT, using knowledge from 7 syntactic, semantic and pragmatic sources.</Paragraph>
    <Paragraph position="4"> In what follows, we first present some related approaches on WSD for MT, focusing oh their limitations (Section 2). We then give some basic concepts on Inductive Logic Programming and describe our approach (Section 3). Finally, we present our initial experiments and the results achieved (Section 4).</Paragraph>
  </Section>
class="xml-element"></Paper>