A CONCEPT SENSE DISAMBIGUATION ALGORITHM FOR CONCEPT MAPS

Concept maps are a graphically rich tool for representing knowledge in natural language. An important aspect for their automatic or semi-automatic processing, including concept mapping, formalization and evaluation, is the identification of the most rational sense of the concepts. In this paper, we present an algorithm for concept sense disambiguation based on contextual analysis, domain information and gloss. The algorithm takes advantage of the whole map's topology and use WordNet as sense repository. Results of preliminary experimental evaluations of the concept disambiguation algorithm applied to several concept maps in the Spanish language are presented and compared with the state of the art.