摘要
针对当前存在的异源POI融合效率低、融合精准度不高及融合成果信息完善程度不足的问题,该文综合考虑异源POI间属性字段信息完整性类型,提出一种多权组与双半径结合的数据融合方法。在设定的匹配半径范围内,依据POI空间位置属性及非空间位置属性相似度,综合计算异源POI间相似度,精准匹配异源重复数据,互补完善POI属性字段信息集。选取郴州市苏仙、北湖区内的百度、腾讯POI进行融合测试,长沙市雨花区内的百度、腾讯、高德POI进行两两融合检验。测试及检验区域异源POI数据,均能在低耗时内实现良好的匹配融合效果。实验结果表明,该方法实现了异源POI数据精准高效融合,具备较强适用性,能够有效丰富空间位置数据,可为地理信息研究与应用提供有力的数据支持服务。
Aiming at the current problems of low efficiency of heterogenous POI fusion,low precision of fusion and insufficient degree of perfecting information of fusion results,this paper proposes a data fusion method combining multi-group weights and double radius by comprehensively considering the type of completeness of information of attribute fields between POIs of heterogenous sources.Within the specified matching radius,according to the similarity of POI spatial location attributes and non-spatial location attributes,the similarity between POIs of different sources is comprehensively calculated,the duplicate data of different sources are accurately matched,and the information set of POI attribute fields is improved in a complementary way.Baidu and Tencent POIs within Suxian and Beihu districts in Chenzhou City are selected for fusion testing,and Baidu,Tencent and AutoNavi POIs within Yuhua District in Changsha City are verified for mutual fusion.The domain heterogeneous source POI data in the test area and the verification area all achieve good matching and fusion effect in low time consumption.The experimental results show that the method achieves accurate and efficient fusion of heterogeneous POI data with strong applicability,effectively enriches spatial location data,and can provide powerful data support services for geographic information research and application.
作者
钟顺杰
葛小三
ZHONG Shunjie;GE Xiaosan(School of Surveying and Land Information Engineering,Henan Polytechnic University,Jiaozuo 454003,Henan,China)
出处
《测绘科学》
CSCD
北大核心
2023年第11期230-244,共15页
Science of Surveying and Mapping
基金
国家自然科学基金项目(41572341)
河南省自然科学基金项目(222300420450)
关键词
异源POI
多权组
双半径
综合相似度
融合
heterogenic POI
multi-weight group
double radius
comprehensive similarity
fusion
