The paper presents a method for word sense disambiguation based on parallel corpora. The method exploits recent advances in word alignment and word clustering based on automatic extraction of translation equivalents and being supported by available aligned wordnets for the languages in the corpus. The wordnets are aligned to the Princeton Wordnet, according to the principles established by EuroWordNet. The evaluation of the WSD system, implementing the method described herein showed very encouraging results. The same system used in a validation mode, can be used to check and

This paper describes techniques for unsupervised word sense disambiguation of English and German medical documents using UMLS. We present both monolingual techniques which rely only on the structure of UMLS, and bilingual techniques which also rely on the availability of parallel corpora. The best results are obtained using relations between terms given by UMLS, a method which achieves 74% precision, 66% coverage for English and 79% precision, 73% coverage for German on evaluation corpora and over 83% coverage over the whole corpus. The success of this technique for German shows that a lexical resource giving relations between concepts used to index an English document collection can be used for high quality disambiguation in another language.

Abstract. Word Sense Disambiguation (WSD from now on) represents an established task within Natural Language Processing community, aiming at finding the right sense of a word occurring in a free running text through the use of a computer algorithm. Currently, most of the WSD approaches consider only monolingual texts, and, as such, they rely mainly on the discriminatory power of the words appearing in the same context with the targeted words (the words to be disambiguated). We propose a more precise WSD method, based on parallel texts, and relying on interlingually aligned wordnets. The method exploits recent advances in translation equivalents extractions and word alignment. 1.

Word sense disambiguation is a core problem in many tasks related to language processing. In this paper, we introduce the notion of soft word sense disambiguation which states that given a word, the sense disambiguation system should not commit to a particular sense, but rather, to a set of senses which are not necessarily orthogonal or mutually exclusive. The senses of a word are expressed by its WordNet synsets, arranged according to their relevance. The relevance of these senses are probabilistically determined through a Bayesian Belief Network. The main contribution of the work is a completely probabilistic framework for wordsense disambiguation with a semi-supervised learning technique utilising WordNet.

Second Language Information Transfer in Automatic Verb Classification -- A Preliminary Investigation Vivian Tsang Master of Science Graduate Department of Computer Science University of Toronto 2001 Lexical semantic classes incorporate both syntactic and semantic information about verbs. Lexical semantic classification of verbs provide a great deal of useful information about the possible usage of each verb. In our work, we explore the use of multilingual corpora in the automatic learning of verb classification. We extend the work of Merlo and Stevenson (2001a), in which statistics on simple syntactic features extracted from textual corpora were used to train an automatic classifier for three lexical semantic classes of English verbs. We hypothesize that some lexical semantic features which are di#cult to detect superficially in English may manifest themselves syntactically in another language. In our two-way classification task, features from multiple languages achieve an accuracy as high as 81%, making a small bitext a useful alternative to using a large monolingual corpus for verb classification. In this thesis, experimental results are presented and future extensions are discussed.

This paper describes techniques for unsupervised word sense disambiguation of English and German medical documents using UMLS. We present both monolingual techniques which rely only on the structure of UMLS, and bilingual techniques which also rely on the availability of parallel corpora.

It is generally agreed that the ultimate goal of research into Word Sense Disambiguation (WSD) is to provide a technology which can benefit applications; however, most of the work in this area has focused on the development of application-independent models. Taking Machine Translation as the application, we argue that this strategy is not appropriate, since many aspects of algorithm design, such as the sense repository, depend on the application. We present evidence for this by investigating the disambiguation of nine verbs in English-Portuguese translations.

This paper describes techniques for unsupervised word sense disambiguation of English and German medical documents using the Unified Medical Language System (UMLS). We present both monolingual techniques which rely only on the structure of UMLS, and bilingual techniques which also rely on the availability of parallel corpora. The best results are obtained using relationships between terms given by UMLS, a method which achieves 74% precision, 66% coverage for English and 79% precision, 73% coverage for German on evaluation corpora and over 83% coverage over the whole corpus. The success of this technique for German shows that a lexical resource giving relationships between concepts used to index an English document collection can be used for high quality disambiguation in another language. Document