摘要
针对现有本体映射过程中相似度计算方法的精度及效率的不足,提出一种新的综合概念相似度算法模型,该算法分别对本体概念的名称、属性和实例相似度进行计算,过程中融合了信息增益和聚类集方法,并最终对三种相似度量结果加权综合。实验表明,算法得出的概念相似度计算结果在合理性和准确率上都有所提高。
This paper is aimed at the shortage of similarity calculation method of the accuracy and efficiency ofpresent ontology mapping process. It puts forward a new comprehensive concept similarity algorithm model, and this algorithm calculates the ontology concepts' name, attribute and examples similarity. It's process has integrated the information gain and cluster set method. It finally weights comprehensive to the three kind of similarity metrics. The results show that, this algorithm in concept similarity calculation efficiency and accuracy are improved.
出处
《齐齐哈尔大学学报(自然科学版)》
2013年第4期18-22,共5页
Journal of Qiqihar University(Natural Science Edition)
基金
国家自然科学基金资助项目(51174257)
关键词
本体映射
概念相似度
聚类集
信息增益
ontology mapping
concept similarity
cluster aggregation
information gain