基于混合采样的城市功能区识别系统研究  

Research on identification system of urban functional area based on mixed sampling

作　　者：吴俊杰 魏山山 WU Junjie;WEI Shanshan(College of Computer Science and Technology,Taiyuan Normal University,Yuci Shanxi 030619,China)

机构地区：[1]太原师范学院计算机科学与技术学院,山西榆次030619

出　　处：《智能计算机与应用》2024年第1期185-190,共6页Intelligent Computer and Applications

摘　　要：随着城市功能分区研究在时空尺度上不断细化,多源数据融合有利于推动城市功能分区研究的精细化发展,但多源数据存在数据不平衡现象,本文通过混合采样算法,减少数据不平衡带来的影响。通过空间面积和公众认知度对兴趣点(POI)数据进行权重赋值,利用K近邻和潜在狄利克雷(Latent Dirichlet Allocation,LDA)语义分析对两类数据构建数据集,分别对两类数据进行频数密度分析,对多源数据进行加权平均特征融合,将融合后的数据按频数密度差值的方法划分单一和混合功能区,并在ArcGIS平台渲染展示,从而实现对城市功能区可视化识别划分。利用高德地图与功能区识别结果进行精度验证,结果表明:应用该方法能快速及有效地识别城市功能区,功能区识别总体精度为86.6%,证明该方法现实可行,可为城市未来发展规划与管理提供借鉴。With the continuous refinement of urban functional zoning research on the temporal and spatial scale,multi-source data fusion is conducive to promoting the refined development of urban functional zoning research.However,there is data imbalance in multi-source data,and this paper uses hybrid sampling algorithms to reduce the impact of data imbalance.The POI data is weighted by spatial area and public awareness,the two types of data are clustered and analyzed by K-nearest neighbor and LDA semantic analysis,the frequency density analysis of the two types of data is carried out respectively,and then the weighted average feature fusion of multi-source data is carried out,and the fused data is divided into single and mixed functional areas according to the method of frequency density difference,and rendered and displayed on the ArcGIS platform,so as to realize the visual identification and division of urban functional areas.The accuracy verification is carried out by using the Amap and the functional area recognition results.The results show that the application of this method can quickly and effectively identify urban functional areas,and the overall accuracy of functional area recognition is 86.6%,which proves that the proposed method is realistic and feasible and can provide reference for future urban development planning and management.

关 键 词：POI 数据不平衡 频数密度 城市功能区 

分 类 号：TP399[自动化与计算机技术—计算机应用技术]