基于非线性映射的城市化水平聚类分析

采用非线性映射分析方法对区域城市化水平进行了聚类分析.结果表明,非线性映射方法通过对高维样本进行降维处理,使不同城市化水平等级的城市得以清晰地划分,结果直观形象地显示在二维平面中,从而实现了水平近似城市的聚类,克服了以往系统聚类分析法所得结果不够直观、存在虚假归并等问题,为城市化水平的定量判定提供了一种新方法.

The Clustering Analysis for the Urbanization Level Based on Nonlinear Mapping Analysis

The urbanization levels of districts are evaluated by the method of nonlinear mapping analysis.The results are visually showed in two-dimensional surface.The map obtained by nonlinear mapping analysis could well represent the classification of different cities with different levels and the results are much better than that obtained by system cluster analysis.The nonlinear mapping method overcome the shortcomings such as not showing the result directly and clustering cities incorrectly.A new and reasonable approach to evaluate the urbanization level is developed.