地理研究 ›› 2016, Vol. 35 ›› Issue (4): 703-716.doi: 10.11821/dlyj201604009

• 研究论文 • 上一篇    下一篇

基于POI数据的广州零售商业中心热点识别与业态集聚特征分析

陈蔚珊1(), 柳林1,2, 梁育填1()   

  1. 1. 中山大学地理科学与规划学院,综合地理信息研究中心,广州510275
    2. Department of Geography, University of Cincinnati, Cincinnati, OH 45221-0131, USA
  • 收稿日期:2015-12-20 修回日期:2016-03-14 出版日期:2016-04-20 发布日期:2016-04-27
  • 作者简介:

    作者简介:陈蔚珊(1981- ),女,广东汕头人,博士研究生,主要研究方向为城市商业地理。E-mail: shanshan8108@163.com

  • 基金资助:
    国家自然科学基金项目(41301112);国家863项目(2013AA122302);广东省教育厅科技项目(2013KJCX0006)

Retail center recognition and spatial aggregating feature analysis of retail formats in Guangzhou based on POI data

Weishan CHEN1(), Lin LIU1,2, Yutian LIANG1()   

  1. 1. School of Geography and Planning, Center of Integrated Geographic Information Analysis, Sun Yat-sen University, Guangzhou 510275, China
    2. Department of Geography, University of Cincinnati, Cincinnati, OH 45221-0131, USA
  • Received:2015-12-20 Revised:2016-03-14 Online:2016-04-20 Published:2016-04-27

摘要:

商业中心是城市零售活动的重要载体,优化商业资源在城市内部空间的合理配置,摸清不同零售经营形态的区位选择,显得尤为重要。以面向公众服务的商业机构兴趣点(POI)数据为研究对象,提出一种城市商业中心与零售业态集聚区识别的方法;以广州市为例,分析商业活动的热点地区以及零售业态集聚区的空间分布特征。研究表明:① 根据核密度估计法提取的商业中心在等级上表现出由城市中心圈层向外围圈层扩散的趋势,结果符合客观事实。② 以街区为单元,商业网点密度符合局域Getis-Ord G*指数统计特征的热点区域主要分布在越秀区和天河区,广州市零售业发展的双核心空间格局已经形成。③ 不同的零售业态对商业集聚的区位选择具有显著差异性,百货商店、超市、便利店等零售经营形态的空间集聚特征与该业态的市场定位、经营模式及选址策略基本吻合。总体来看,基于POI数据的广州零售业集聚空间分析结果能够反映实体零售企业行为与广州商业经济分布的相关性,有助于提高政府部门商业规划和零售商选址前期研究的客观性和科学性。

关键词: POI数据, 商业中心识别, 零售业态, 集聚特征, 广州

Abstract:

It is important to get a clear understanding of rational allocation of commercial resources and spatial aggregating feature of retail formats in a metropolis. The use of big data for retailing agglomeration analysis has already became a new trend of commercial quantitative research, yet little attention has been given to the hotspots recognition of retail formats and their cluster characters so far. In this paper, an approach of urban retail center recognition based on POI (Point of Interest) data are proposed. With Guangzhou for a case study, we analyze the hotspots of retail activities and inner aggregating distribution of retail formats by using the method of KDE (Kernel Density Estimation) and Getis-Ord G*. The results show that: (1) The commercial levels of retail centers which are extracted by KDE method have a strong correlation with the density distribution. The density values of retail centers are decreasing from urban center to the suburbs, which shows the trend of commercial grade within the city is diffusing from center to periphery. This result is in line with the objective facts. (2) The retailing hotspots are mainly distributed in Yuexiu and Tianhe districts, which proves that dual-core commercial space has taken shape in Guangzhou. (3) The spatial aggregating feature of retail formats (including department stores, shopping centers, supermarkets, convenience stores) are quiet different. Department stores and shopping centers are clustered around the retail centers. Supermarkets are mainly distributed in periphery of downtown area. And convenience stores are only centralized in hotspots in business locations. From the results of spatial analysis, the aggregating features of retail formats are consistent with retailers' operation and their location strategy. Overall, the results of retailing distribution analysis based on the POI data can explain part of the corporate behaviors difference and commercial economy distribution within urban areas. The study results of retailing activities are also conducive to the strategy-making process of both governments and retailers.

Key words: POI data, retailing hotspots recognition, retail formats, aggregating feature, Guangzhou