Regresar

Semantic Processing Method to Improve a Query-Based Approach for Mining Concept Maps

Abstract:

Concept maps are powerful visual tools for organizing and representing knowledge. The development of computational systems for building and managing concept maps has facilitated the creation of concept maps repositories, which can be considered valuable knowledge models. In this paper, a semantic processing method to improve a query-based approach for mining concept maps repositories is proposed. The method is based on a mechanism of semantic extension of the represented concepts in the concept maps and a set of rules for guiding the semantic integration of the conceptual structures and retrieval information processes. WordNet is used for the semantic extension of concepts, which is supported in a disambiguation algorithm. Results of experimental evaluations of the disambiguation algorithm applied to several concept maps are presented and compared with the state of the art. The application of the proposed method is illustrated through a study case with promising results.