An Innovative High-Dimensional Clustering Algorithm and its Application to Urban Stratification and Planning
Zhuohao Du, Yanshan Qian, Chuanan Li, Yunqi Hu, Junda Qiu · 2024
With the acceleration of urbanisation, efficient and scientific urban planning and policy making is particularly important. However, traditional clustering algorithms tend to segment data based on a single objective or parameter in the process of urban stratification, which is easy to ignore the complexity and dynamics within the city when dealing with multidimensional and dynamically changing urban data, resulting in less accurate and comprehensive clustering results. This paper introduces an innovative clustering algorithm based on the Pareto optimal theory, which effectively improves the accuracy of clustering and the ability of multi-dimensional data processing by selecting the Steiner point as the clustering centre. Applying this algorithm to urban stratification and planning can not only reveal the development situation of the city in a more comprehensive way, but also provide the government with more scientific and detailed data support, which can promote the formulation of more reasonable and sustainable urban development strategies.