提出一种新的本体集成方法。分析了本体集成的原因,阐述了本体集成时应遵循的4条基本原则,并给出了集成的分类,提出了一种基于OWL DL图闭包的本体集成方法。该方法将OWL DL本体抽象为RDFS图模型,根据给定的OWL DL推理规则生成OWL DL本体的图闭包,在此基础上进行本体集成,同时提出了几种计算实体相似度的方法,将本方法与COMA++和FCA-merge进行实验对比,本方法在准确率和召回率上占优势。
A new method for ontology integration was proposed. Firstly, after having analyzed the reasons of ontology integration, four basic rules were put forward and the classifications for it were given. Secondly, a method based on the closure of OWL DL for ontology integration was proposed. In this method, OWL DL ontologies were abstracted as RDFS graph models. Using the OWL inference roles, the closure of OWL DL ontologies could be constructed. Based on these closures, the integration could be carried out. During this part, some measure methods for similarity between entities were proposed too. Lastly, for comparing this method with COMA ++ and FCA-merge, experiments were done. Results show that the method is better than COMA + + and FCA-merge in precision and recall.