Research on Multi-Source Data Fusion Urban Functional Area Identification Method Based on Random Forest Model

Identifying urban functional zones is an important task in urban planning and smart city construction. Accurately identifying urban functional zones and analyzing their spatial distribution is crucial for optimizing urban spatial layout, improving urban management and balancing human–environment interaction. However, most of the existing studies focus on the analysis of individual data sources, which have difficulty fully reflecting the complex spatial structures and functional differences in cities. To solve this problem, this paper proposes a new method of urban functional area identification which integrates multi-source data and advanced algorithms. By clustering the results of multi-factor weighted kernel density, the paper can more accurately quantify the spatial distribution characteristics of urban functional areas and better reflect the functional differences among urban areas. At the same time, this paper uses the Random Forest model to optimize the POI data and the building data to improve the classification accuracy and the generalization ability of the model. The results show that the distribution of functional areas in the Fifth Ring Road region of Beijing presents the characteristics of diversification and agglomeration: the core urban area is dominated by high-density commercial service and public service functional areas, with a high degree of functional integration; The peripheral areas are mainly residential areas and green areas, with dispersed distributions, but clear functions. The overall accuracy reaches 87%, and the model can effectively reflect the distribution and spatial characteristics of urban functional areas.