大数据背景下基于兴趣点（POI）数据的城市功能区识别——以南京市主城区为例

Identification of Urban Function Area Based on Point of Interest （POI） Date in the Big Data Background——A Case Study of Nanjing Downtown

本文以南京市主城区的322000条POI数据为基础,利用基于知识的决策树分类方法对南京市的城市功能区进行识别,探索城市功能类型.结果显示1.南京市都市区构建形成以主城为核心,以放射性交通走廊为发展轴,以生态空间为绿楔的现代都市区空间格局.2.南京市结合自然条件、交通走廊布局和区域联系主要方向,都市区内形成了“一带五轴”的城镇布局结构.3.与南京市城市总体规划进行对比分析,显示基于POI数据的城市功能区识别结果精度较高,与规划分区基本吻合.为城市规划的编制提供了实际的参考依据,且为利用开放数据进行规划应用提供了新的思路和方法.

Based on the 32200 POI data of Nanjing downtown, this paper using decision tree classifcation method to identify urban function areas in Nanjing, and to explore urban functional types. This result shows that, frst, the metropolitan area of Nanjing is formed the spatial pattern of modern metropolitan area, that is composed of one core by downtown, the radioactive traffc corridor as the development axis, and the ecological space as the green wedge. Second, Combining the natural conditions, the layout of transportation corridors and the main directions of regional relations, Nanjing is formed a ＂One Belt and fve Axis＂ urban layout structure in the urban area. Third, compared with the urban master plan of Nanjing, the result shows that the accuracy of the urban functional area identifcation based on POI data is high, which is basically consistent with the planning division. This method provides actual reference for urban planning compilation, and new ideas and methods for the application of planning with open data.