A hierarchical spatial unit partitioning approach for fine‐grained urban functional region identification

Abstract

Urban functional regions (UFRs) are formed and developed with human social actions and can reflect urban land use types. Appropriately identifying UFRs helps solve existing urban problems, optimize the spatial structure of cities, and provide a database for sustainable urban development. Most existing studies focus on developing novel methods and fusing multiple data sources, but neglect the impact of heterogeneous spatial units on UFR identification results. In this work, a hierarchical spatial unit partition approach was proposed to split the research area into many hierarchical units whilst considering the mixed degree of each unit. We further explored its capacity for identifying UFRs by integrating it with three widely used UFR identification methods. The results reveal that our proposed approach can correctly identify 10% more UFRs compared with the traditional grid‐based method, showing the efficiency of our proposed approach for identifying UFRs at a finer scale. This work provides support for those who are engaged in urban planning and urban policymaking, promoting urban sustainable development.

Identification and spatio-temporal characterization of urban functional areas based on POI data

“…With the rapid development of social economy and the gradual acceleration of urbanization, the urban spatial structure is constantly changing. At present, the division of urban functional areas mainly adopts traditional methods such as subjective discrimination and survey statistics [1,2] . The emergence of point-of-interest (POI) data has provided new data sources and methods for urban functional area identification research [3] , POI data are now widely used in hotspot analysis of urban facilities [4] , urban economic development [5] , urban spatial research [6] , urban functional area identification [7], and other research hotspot problems.…”

“…The determination of weights has become one of the ways to improve the accuracy of urban functional area identification. Jiang et al used public awareness and spatial area as influencing factors to determine the weights of POI data [2] ; Hu et al proposed the identification method based on frequency density and POI type ratio [8] . With the development of machine learning, Zhao et al introduced random forest to assign weights to POI data to analyze the production-life-ecology spatial pattern of Zhengzhou [9] .…”

A Multi-Feature Fusion Method for Urban Functional Regions Identification: A Case Study of Xi’an, China

“…Some have proposed a hierarchical spatial unit partitioning method, dividing the research area into many hierarchical units while considering the degree of mixture in each unit. At a finer scale, these research methods and improvements mentioned above further enhance the efficiency of identifying urban functional regions [39].…”

Semantic-Enhanced Graph Convolutional Neural Networks for Multi-Scale Urban Functional-Feature Identification Based on Human Mobility

“…In this research field, one of the most longstanding and far-reaching geographic issues is the modifiable areal unit problem (MAUP), concerned with the spatial scale and the zoning sub-problem [3,4]. Moreover, appropriate spatialunit zoning is significant for urban planning, urban governance, and synthetic studies of urban geography [5]. Given the spatial heterogeneity of urban spatial units, predicting unknown functional features of spatial units from the effective representation of units' relationships is treated as an extensive challenge in urban science [6].…”

“…One of the major reasons is the limitation of the range and contents of their datasets. The deliberate arrangement of spatial units will considerably increase the prediction accuracy in GCNN-based deep-learning tasks; however, the spatial scale effect has not been validated in the identification of urban land use from spatial interaction with graph embedding [5]. In addition, due to the deficiency of transparency for most deep learning models, explainability continues to be a major concern in attaining greater insight into the characteristics underlying mobility flows [27].…”