Multilingual Word Sense Disambiguation Using Wikipedia

International Joint Conference on Natural Language Processing, pages 498–506, Nagoya, Japan, 14-18 October 2013. Multilingual Word Sense Disambiguation Using Wikipedia Bharath Dandala Dept. of Computer Science University of North Texas Denton, TX BharathDandala@my.unt.edu Rada Mihalcea Dept. of Computer Science University of North Texas Denton, TX rada@cs.unt.edu Razvan Bunescu School of EECS Ohio University Athens, OH bunescu@ohio.edu Abstract We present three approaches to word sense disambiguation that use Wikipedia as a source of sense annotations. Starting from a basic monolingual approach, we de- velop two multilingual systems: one that uses a machine translation system to cre- ate multilingual features, and one where multilingual features are extracted primar- ily through the interlingual links available in Wikipedia. Experiments on four lan- guages confirm that the Wikipedia sense annotations are reliable and can be used to construct accurate monolingual sense clas- sifiers. The experiments also show that the multilingual systems obtain on average a substantial relative error reduction when compared to the monolingual systems. 1 Introduction and Motivation Ambiguity is inherent to human language. In particular, word sense ambiguity is prevalent in all natural languages, with a large number of the words in any given language carrying more than one meaning. For instance, the English noun plant can mean green plant or factory; similarly the French word feuille can mean leaf or paper. The correct sense of an ambiguous word can be se- lected based on the context where it occurs, and correspondingly the problem of word sense disam- biguation is defined as the task of automatically assigning the most appropriate meaning to a poly- semous word within a given context. Two well studied categories of approaches to word sense disambiguation (WSD) are repre- sented by knowledge-based (Lesk, 1986; Galley and McKeown, 2003; Navigli and Velardi, 2005) and data-driven (Yarowsky, 1995; Ng and Lee, 1996; Pedersen, 2001) methods. Knowledge- based methods rely on information drawn from wide-coverage lexical resources such as WordNet (Miller, 1995). Their performance has been gen- erally constrained by the limited amount of lexi- cal and semantic information present in these re- sources. Among the various data-driven WSD methods proposed to date, supervised systems have been observed to lead to highest performance in the Senseval evaluations 1. In these systems, the sense disambiguation problem is formulated as a super- vised learning task, where each sense-tagged oc- currence of a particular word is transformed into a feature vector which is then used in an automatic learning process. Despite their high performance, the supervised systems have an important draw- back: their applicability is limited to those few words for which sense tagged data is available, and their accuracy is strongly connected to the amount of available labeled data. In this paper, we address the sense-tagged data bottleneck problem by using Wikipedia as a source of sense annotations. Starting with the hy- perlinks available in Wikipedia, we first generate sense annotated corpora that can be used for train- ing accurate and robust monolingual sense clas- sifiers (WIKIMONOSENSE, in Section 2). Next, the sense tagged …

Content truncated to 500 words