DBSCAN Spatial Clustering Analysis of Urban “Production–Living–Ecological” Space Based on POI Data: A Case Study of Central Urban Wuhan, China

Abstract

As urban spatial patterns are the prerequisite and foundation of urban planning, spatial pattern research will enable its improvement. The formation mechanism and definition of an urban “production–living–ecological” space is used here to construct a classification system for POI (points of interests) data, crawl POI data in Python, and DBSCAN (density-based spatial clustering of application with noise) to perform cluster analysis. This mechanism helps to determine the cluster density and to study the overall and component spatial patterns of the “production–living–ecological” space in the central urban area of Wuhan. The research results are as follows. (1) The spatial patterns of “production–living–ecological” space have significant spatial hierarchical characteristics. Among them, the spatial polarizations of “living” and “production” are significant, while the “ecological” spatial distribution is more balanced. (2) The “living” space and “production” space noise points account for a small proportion of the total and are locally clustered to easily become areas with development potential. The “ecological” space noise points account for a large proportion of the total. (3) The traffic accessibility has an important influence on the spatial patterns of “production–living–ecological” space. (4) The important spatial nodes of each element are consistent with the overall plan of Wuhan, but the distribution of the nodes for some elements is inconsistent. The research results show that the POI big data can accurately reveal the characteristics of urban spatial patterns, which is scientific and practical and provides a useful reference for the sustainable development of territorial and spatial planning.