基于百科词条的本体概念聚类方法研究

该文面向本体关系集合的自动构建，提出一种基于百科词条的本体概念聚类方法，用于发现领域概念之间的语义关系。在给定领域本体概念集合的条件下，该方法首先获取相关的百科词条并建立每一概念的向量模型，然后根据距离判别法进行概念聚类，得到概念间的相近关系。采用该方法对3个领域中的领域概念集合进行聚类，实验结果表明，该文方法比传统聚类算法有更好的聚类结果，有助于概念间关系的自动获取和领域本体自动构建。

Ontology Concepts Clustering Based on Encyclopedia Entr

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.