摘要
本体相似度计算是本体映射的关键环节。本体的实例、关系、属性、结构等信息是相似度计算需要考虑的重要因素。针对目前本体映射过程中相似度计算所存在的问题,提出了一种综合的相似度计算方法。首先判断不同本体之间是否存在相关性。若相关,则充分考虑各种相关因素,从语义和概念两个层面来进行比较,然后给出了本体的综合相似度计算方法。最后采用了两组测试数据对该方法进行实验,并与GLUE系统的概率统计方法进行了实验对比。实验结果表明,该方法能够有效确保相似度计算的准确性。
Similarity computation among ontologies is the critical tache in the process of mapping. The information about instances,relations,sturctures and attributes are also important factors. Aiming at the current problems, put forward a compositive approach of similarity computation. The relativity among different domain ontologies was judged first. And then in this subsumption, based on semantic level and concept level, a comprehensive similarity measuring method was proposed, after taking a full consideration about relative factors. This method was tested with two datasets and compared with probability statistical method of GLUE system. Experimental results indicate that this approach can significantly improve the precision.
出处
《计算机科学》
CSCD
北大核心
2008年第12期142-145,182,共5页
Computer Science
基金
国家自然科学基金(60773100)
教育部科学技术研究重点项目(205014)
河北省教育厅科研计划项目(2006143)
关键词
本体
本体映射
相关度
相似度
本体相似度
概念相似度
Ontology,Ontology mapping,Ontology relativity,Ontology similarity,Concept similarity
