县域城镇化水平差异的人工神经网络测定——以陕西省为例

在已有研究的基础上,运用人工神经网络的理论和方法,构建了BP神经网络模型,并对2008年陕西省83个县的县域城镇化水平进行了测度。结果表明:陕西省县域城镇化水平存在显著分异,评价结果与专家的判断基本一致。陕北、关中地区县域城镇化水平较高,陕南则相对较低,各自区内差异明显。可见,BP神经网络模型运用于县域城镇化水平简单、实用,且有效避免了人工赋权的主观性,具有良好的应用前景。

Examination to the differentiations in urbanization level of regional counties by using artificial neural networks:a case study of Shanxi Province

Based on BP neural network model and GIS technology,this article makes a comprehensive assessment on the county-level urbanization in Shaanxi Province.The results show that the characteristics of county-level urbanization are convex distribution.In Guanzhong area counties are also characterized with a high comprehensive development level and less regional differences.In the north of Shanxi Province,counties with high level urbanization development appear to cluster.In the south of Shanxi Province,the levels of the comprehensive development of most counties are low.This article also reveals that BP neural network model applied to the county-level urbanization is simple,practical,but an effective way to avoid the subjectivity of human empowerment and has a good prospect.