Abstract:
To build the set of ontology relations automatically, this paper presents a preliminary study on a concept clustering method based on encyclopedia entries of obtaining semantic relations among concepts. Given a set of domain ontology concepts, this method clusters concepts in 3 steps: 1) obtaining encyclopedia entries; 2) modeling each of the concepts into a vector; 3) clustering concepts using the distance discrimination method. Clustering experiments on 3 sets of domain ontology concepts demonstrate that the proposed method shows better results compared with classical clustering methods and has good potentials for identifying related concepts automatically in the ontology building tasks.