基于粗糙集-神经网络的城市产业生命周期识别

以城市经济为背景,提出了基于粗糙集-RBF神经网络的城市产业生命周期识别方法.首先运用基于MDV函数与信息熵的模糊聚类算法进行连续属性离散化处理,然后采用粗糙集理论约简出重要指标体系,最后将训练样本输入RBF神经网络进行学习和训练,并对检验样本的产业生命周期阶段进行判断.对大连市669组样本产业的分析结果表明:基于MDV函数与信息熵的模糊聚类算法能够有效改善离散化效果,且与通常采用的模糊评价法相比,该方法对检验样本预测精度更高,是一种有效和实用的城市产业生命周期识别工具.

Identifying method of city's industry life cycle based on integration of rough sets and neural network

Based on integration of rough sets and neural network an identifying method of city's industry life cycle is proposed.Firstly continuous attribute values are discretized by using the fuzzy clustering algorithm based on maximum discernibility value(MDV) function and information entropy.Then the attributes are reduced by rough sets.At last,the RBF neural network is trained with training samples and the industry life cycle stages of testing samples ale identified.The analysis resuits taked 669 industries of Dalian city as samples show that the fuzzy clustering algorithm Call improve discretization performance effectively.Compared with the usual fuzzy assessment method the forecasting precision of the integration method is higher,and it is an efficient and practical tool to identify life cycle of city's industry.