Using Relevant Domains Resource for Word Sense Disambiguation

This paper presents a new method for Word Sense Disambiguation based on the WordNet Domains lexical resource [4]. The underlaying working hypothesis is that domain labels, such as ARCHITECTURE, SPORT and MEDICINE provide a natural way to establish semantic relations between word senses, that can be used during the disambiguation process. This resource has already been used on Word Sense Disambiguation [5], but it has not made use of glosses information. Thus, we present in first place, a new lexical resource based on WordNet Domains glosses information, named " Relevant Domains ". In second place, we describe a new method for WSD based on this new lexical resource (" Relevant Domains "). And finally, we evaluate the new method with English all­words task of SENSEVAL­2, obtaining promising results.