Word sense disambiguation based on gloss expansion

Dare to claim that the present age is the age of information and the need to use machine for processing and translating texts has become an undeniable need. One of the most important tasks which mostly affect the accuracy and quality of machine translation is word sense disambiguation (WSD) or choosing an appropriate sense for words that have multiple meanings. There are different views and a variety of methods that have been proposed for WSD. In this paper, we try to take advantage of supplementary information on ambiguous words included in the context and provide a scoring method to resolve the ambiguity in a knowledge based manner. For this purpose, we gather two lists of words; From WordNet, we extract related words of a sense and from the corpus, we select the words surrounding the target word. Finally, we achieved 80.1% accuracy by applying the proposed method on TWA corpus, which is promising compared to the results obtained by similar methods in the literature.