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 th...

Supervised approaches to Word Sense Disambiguation (WSD) have been shown to outperform other approaches but are hampered by reliance on labeled training examples (the data acquisition bottleneck). This paper presents a novel approach to the automatic acquisition of labeled examples for WSD which makes use of the Information Retrieval technique of relevance feedback. This semi-supervised method ...

Domain adaptation (DA), which involves adapting a classifier developed from source to target data, has been studied intensively in recent years. However, when DA for word sense disambiguation (WSD) was carried out, the optimal DA method varied according to the properties of the source and target data. This paper describes how the optimal method for DA was determined depending on these propertie...

There are many similarity measures to determine the similarity relatedness between two words. Measures of similarity or relatedness are used in such applications as word sense disambiguation. One of the methods used to resolve WSD is the Lesk algorithm. The performance of this algorithm is connected with the similarity relatedness between all words in the text, i.e the success rate of WSD shoul...

In this paper, we discuss the problem of Word Sense Disambiguation (WSD) and one approach to solving the lexical sample problem. We use training and test data from SENSEVAL-3 and implement methods based on Naı̈ve Bayes calculations, cosine comparison of word-frequency vectors, decision lists, and Latent Semantic Analysis. We also implement a simple classifier combination system that combines the...

In this paper, we improve an unsupervised learning method using the ExpectationMaximization (EM) algorithm proposed by Nigam et al. for text classification problems in order to apply it to word sense disambiguation (WSD) problems. The improved method stops the EM algorithm at the optimum iteration number. To estimate that number, we propose two methods. In experiments, we solved 50 noun WSD pro...

Beside the word order problem, word choice is another major obstacle for machine translation. Though phrase-based statistical machine translation (SMT) has an advantage of word choice based on local context, exploiting larger context is an interesting research topic. Recently, there have been a number of studies on integrating word sense disambiguation (WSD) into phrase-based SMT. The WSD score...

This paper studies the importance of qualia relations for Word Sense Disambiguation (WSD). We use a graph-based WSD algorithm over the Italian WordNet and evaluate it when adding different kinds of qualia relations (agentive, constitutive, formal and telic) taken from PAROLE-SIMPLE-CLIPS (PSC), a Language Resource based on the Generative Lexicon theory. Some qualia relations, specially telic, a...

The goal of the Cross-lingual Word Sense Disambiguation task is to evaluate the viability of multilingual WSD on a benchmark lexical sample data set. The traditional WSD task is transformed into a multilingual WSD task, where participants are asked to provide contextually correct translations of English ambiguous nouns into five target languages, viz. French, Italian, English, German and Dutch....