面向地下管线空间数据匹配的概念语义相似性方法  

A Concept Semantic Similarity Method for Underground Pipeline Spatial Data Matching

作　　者：房彩丽[1,2] 张书亮[1,3] Fang Caili;Zhang Shuliang(Key Laboratory of Virtual Geographic Environment for the Ministry of Education, Nanjing Normal University, Nanjing 210023;College of Computer Information Engineering , Henan University, Kaifeng 475004;Jiangsu Center for Collaborative Innovation in Geographical Information Resource Development and Application, Nanjing 210023)

机构地区：[1]南京师范大学虚拟地理环境教育部重点实验室,南京210023 [2]河南大学计算机与信息工程学院,开封475004 [3]江苏省地理信息资源开发与利用协同创新中心,南京210023

出　　处：《计算机辅助设计与图形学学报》2017年第4期720-727,共8页Journal of Computer-Aided Design & Computer Graphics

基　　金：国家自然科学基金(41171301)

摘　　要：正针对城市地下综合管线和专业管线空间数据匹配中语义表达不足、语义相似性计算方法简单及匹配质量不高等问题,提出一种顾及管线空间特征的概念语义相似性方法.首先列举管线概念属性项以表达概念语义,分析管线实体拓扑特征确定实体空间关联情况,融合其他特征建立管线本体;然后从管线信息概念内涵出发,提出基于概念语义的管线空间数据匹配模型,其中利用概念属性、空间特征及本体层次结构匹配指标,结合权重信息计算管线实体相似度大小确定匹配实体.实验结果表明,该方法能更合理地计算管线实体间的相似度,明显提升管线空间数据匹配的质量.Considering the shortcomings in the current methods for spatial data matching between urban undergroundintegrated and professional pipelines,such as insufficient semantic expression,rough calculation methodfor semantic similarity and low matching quality,this study proposed a novel conceptual semantic similaritymethod that takes account of pipeline spatial feature.Firstly,the conceptual semantics were expressed by listingthe pipeline conceptual attributes,and the entity spatial association of the pipeline was determined by entity topologicalcharacteristics.The pipeline ontologies were built with other characteristics.Secondly,a matching modelwas proposed based on the pipeline conceptual semantics,which takes advantages of matching metrics includingconceptual attributes,spatial features and ontology hierarchical structure.More importantly,the matching entityin the model is determined by calculating the similarity degree of the pipeline entity with the weight information.The experimental results showed that this method can reasonably calculate the similarity of pipeline entities,andsignificantly improve the quality of pipeline spatial data matching.

关 键 词：概念语义 概念属性 相似性 空间数据匹配 

分 类 号：P208[天文地球—地图制图学与地理信息工程]