SOM网络在福建省城市职能分类中的应用

基于人工神经网络(ANN)中自组织特征映射网络(SOM)的聚类功能,提取城市人口和劳动力结构的多项指标,应用MATLAB6.1软件的神经网络工具箱对福建省城市职能进行分类,最终将23个城市分为6类。同时对分类结果进行分析和讨论,指出SOM网络可以避免传统聚类方法难以克服的一些缺点,并利用其强大的学习功能及良好的自组织性、自适应性和鲁棒性为城市体系的相关研究提供高效便捷的信息处理手段。

THE APPLICATION OF SOM ON THE CLASSIFICATION OF FUJIAN CITIES' FUNCTIONS

Based on the clustering function of self-organizing feature map in ANN,the authors select some factors on population and the structure of labor force in Fujian cities to classify the cities'functions by using the neural network toolbox in MATLAB 6.1. Finally, 23 cities are classified into 6 groups. Above that,this paper analyses the result of classification and points out that SOM can avoid some insurmountable shortcoming in the traditional clustering methods. Its strong learning function,excellent self-organizing feature,self-adapting feature and robustness can work effectively on the study about cities system.