基于合并算子的多源面实体匹配  

Multi-source Planar Entity Matching Based on Merging Operator

作　　者：韩立建 钱海忠[1] 王骁[1] 谢丽敏[1] 陈月 HAN Lijian;QIAN Haizhong;WANG Xiao;XIE Limin;CHEN Yue(Institute of Geospatial Information,Information Engineering University,Zhengzhou 450000,China)

机构地区：[1]信息工程大学地理空间信息学院,河南郑州450000

出　　处：《测绘与空间地理信息》2024年第5期79-82,86,共5页Geomatics & Spatial Information Technology

基　　金：河南省杰出青年科学基金(212300410014);国家自然科学基金(42271463)资助。

摘　　要：在地理空间数据更新中,多源数据的差异性备受关注,因之引发的在数据匹配过程中一对多和多对多匹配关系一直是研究的难点问题。本文提出一种基于合并算子的多源数据相似性计算方法,以城市矢量面状建筑物数据为例,首先对两组不同来源面实体数据,综合考虑面实体的位置、大小和形状特征,以加权平均的方法计算实体间总相似度;其次,用合并算子处理待匹配数据后重新进行匹配;最后通过分析匹配结果差异来判断合并算子的有效性。In geospatial data updating,the difference of multi-source data has attracted much attention,and the one to many and many to many matching relationship in the process of data matching has always been a difficult problem.In this paper,we propose a method to compute the similarity of multi-source data based on merging operator.Taking urban vector planar building data as an example,firstly,for two groups of planar entity data from different sources,considering the location,size and shape characteristics of planar en-tities,the total similarity between entities is calculated by weighted average method;secondly,the data to be matched is processed by merging operator and then matched again;finally,the validity of the merging operator is judged by analyzing the difference of the matching results.

关 键 词：实体匹配 城市建筑物 合并算子 

分 类 号：P208[天文地球—地图制图学与地理信息工程]