摘要
城市昼夜人口数据是了解城市活力和城市综合治理的重要依据。本文以郑州市主城区为研究区,结合无线网络数据和兴趣点(Point of Interest,POI)数据构建人口吸引力指标,采用城乡分区策略和随机森林模型开展人口空间化实验,生成100 m格网尺度的城市昼夜人口分布图;之后将结果与公开格网人口数据进行精度对比,并分析研究区昼夜人口分布格局。结果表明:1)融合无线网络数据和POI数据可以提高格网人口精度,能够有效反映城市昼夜人口分布差异;2)研究区人口分布呈现显著的“中心—边郊—乡村”的衰减模式,中心城区白天人口聚集效应高于夜晚,昼夜人口流动方式主要为内部流动;3)本文估算结果的各类误差均小于公开数据集,与真实人口信息更加符合。
Urban daytime and nighttime population data is an important basis for understanding urban vitality and comprehensive urban governance.In this paper,the main urban area of Zhengzhou city is taken as the research area,and a population attraction index is constructed based on WiFi data and POI data.Based on the urban-rural zoning strategy and random forest model,the population spatialization experiment is carried out to generate the urban daytime and nighttime population distribution map at a 100-meter grid scale.Then,the accuracy of the results is compared with the public grid population data,and the population distribution pattern of the study area is analyzed.The results show that:1)Integrating WiFi data and POI data can improve the accuracy of grid population data,and effectively reflect the differences in urban daytime and nighttime population distribution;2)The population distribution in the research area shows a significant decreasing pattern of"center-suburban-rural",with higher daytime population aggregation than nighttime in the central urban area,and the main mode of diurnal and nocturnal population flow is internal migration;3)The estimation errors observed in this paper are notably lower than those reported in the public dataset and demonstrate greater consistency with actual population data.
作者
李帅
徐青
朱新铭
李雪嫚
闫旭强
LI Shuai;XU Qing;ZHU Xinming;LI Xueman;YAN Xuqiang(Geospatial Information Institute,Information Engineering University,Zhengzhou 450001,China)
出处
《测绘与空间地理信息》
2025年第2期92-96,共5页
Geomatics & Spatial Information Technology
关键词
人口空间化
昼夜人口
POI数据
无线网络数据
随机森林
population spatialization
daytime and nighttime population
Point of Interest(POI)data
WiFi data
random forest
