A Probabilistic Model Based on n-Grams for Bilingual Word Sense Disambiguation

Abstract

Word Sense Disambiguation (WSD) is considered one of the most important problems in Natural Language Processing. Even if the problem of WSD is difficult, when we consider its bilingual version, this problem becomes to be much more complex. In this case, it is needed not only to find the correct translation, but this translation must consider the contextual senses of the original sentence (in a source language), in order to find the correct sense (in the target language) of the source word. In this paper we propose a model based on n-grams (3-grams and 5-grams) that significantly outperforms the last results that we presented at the cross-lingual word sense disambiguation task at the SemEval-2 forum. We use a naive Bayes classifier for determining the probability of a target sense (in a target language) given a sentence which contains the ambiguous word (in a source language). For this purpose, we use a bilingual statistical dictionary, which is calculated with Giza++ by using the EUROPARL parallel corpus, in order to determine the probability of a source word to be translated to a target word (which is assumed to be the correct sense of the source word but in a different language). As we mentioned, the results were compared with those of an international competition, obtaining a good performance.