Analysis of urban functional areas based on graph clustering neural networks

Urban functional areas, as an important fundamental component of urban planning, hold significant importance for optimizing urban layouts, resource allocation, and the rational organization of socio-economic activities. Urban functional area division using traditional methods often has limitations, such as relying on a single data source and using division methods that are mostly based on road network units. To address these issues, this paper considers the impacts of Open Street Map road network data, Point of Interest data, nighttime light data, and land use data on functional area identification and proposes a method based on Graph Clustering Neural Networks to analyze and study urban functional areas. To begin with, the basic datasets are utilized to extract the spatial structure and attribute features of urban functional areas. Then, a transportation network model is constructed by combining the basic characteristics of the road network with socio-economic characteristics. Through the use of Graph Clustering Neural Networks and clustering algorithms, urban functional areas can be identified through clustering through the integration of Graph Neural Networks and clustering algorithms. For functional area analysis, Point of Interest data and land use data are then utilized. Finally, the accuracy of the clustering results is verified by comparison with real images. Beijing was chosen as the study area for this research. The Graph Clustering Neural Network model's urban functional area analysis is highly accurate, allowing for precise identification of the city's spatial distribution characteristics, as demonstrated by the results. The analysis of functional areas offers a better understanding of urban characteristics and distribution patterns, providing more scientific and precise references for urban planning. Additionally, it offers strong support for real-world urban management and decision-making, such as traffic planning, land use optimization, and public resource allocation.