本体演化中基于TFOF的波及效应分析  

Analysis for Ripple-Effect Based on TFOF in Ontology Evolution

作　　者：陈晶 刘钊[1,2,3,4] 顾进广[1,2,3,4] 刘宇[1,2,3,4] CHEN Jing;LIU Zhao;GU Jinguang;LIU Yu(School of Computer Science and Technology,Wuhan University of Science and Technology,Wuhan 430065,Hubei,China;Key Laboratory of Intelligent Information Processing and Real-Time Industrial System in Hubei Province,Wuhan University of Science and Technology,Wuhan 430065,Hubei,China;Institute of Big Data Science and Engineering,Wuhan University of Science and Technology,Wuhan 430065,Hubei,China;Key Laboratory of Rich-Media Knowledge Organization and Service of Digital Publishing Content,National Press and Publication Administration,Beijing 100038,Beijing,China)

机构地区：[1]武汉科技大学计算机科学与技术学院,湖北武汉430065 [2]武汉科技大学智能信息处理与实时工业系统湖北省重点实验室,湖北武汉430065 [3]武汉科技大学大数据科学与工程研究院,湖北武汉430065 [4]国家新闻出版署富媒体数字出版内容组织与知识服务重点实验室,北京100038

出　　处：《武汉大学学报（理学版）》2020年第2期197-204,共8页Journal of Wuhan University:Natural Science Edition

基　　金：国家自然科学基金(61673304);国家社会科学基金重大计划(11&ZD189)。

摘　　要：通过定量地评估本体中概念或实体的波及效应,本体的所有者和使用者均能够准确、全面地分析本体演化。现有方法初步量化了本体中的波及效应,但是并未考虑本体中语义关系强弱对于波及效应的影响。针对上述问题,本文提出了基于术语频率与本体频率(term frequency and ontology frequency,TFOF)的定量评估方法,并将其应用于本体演化研究。该方法利用本体中语义关系的出现频率计算关系的权重,从而获得本体所对应的语义关系矩阵,最后运用改进的Floyd-warshall算法计算各个概念或实体的波及效应。实验结果表明,基于TFOF波及效应的评估方法能够更精准地度量本体演化过程中概念或实体的波及效应,且该方法对不同版本中共有结点的度量结果更加稳定。By evaluating the ripple-effect of concepts or entities in the ontology, both the owner and the user of the ontology can accurately and comprehensively analyze the ontology evolution. This paper proposes a quantitative evaluation method based on the TFOF(term frequency and ontology frequency) and applies it to ontology evolution research due to the fact that the existing methods do not consider the influence of the semantic relationship between the ontologies. This paper uses the frequency of the semantic relationship in the ontology to calculate the weight of the relationship to obtain the semantic relation matrix corresponding to the ontology. Finally,the improved Floyd-warshall algorithm is used to calculate the ripple-effect of each concept or entity. The experimental results show that the evaluation method based on the TFOF ripple-effect can measure the ripple-effect of concepts or entities in the evolution process of ontologies with higher accuracy, and the results of the common nodes in different versions are more stable.

关 键 词：本体演化 波及效应 语义关系 TFOF算法 

分 类 号：TP311.1[自动化与计算机技术—计算机软件与理论]