An integrated RS and ANN design method for city's industry development stage identification

作者：Wang De Lu^*; He Xin; Zhao Shen

来源：Journal of Applied Sciences, 2013, 13(16): 3251-3256.

DOI：10.3923/jas.2013.3251.3256

• Continuous attribute
• Fuzzy evaluation method
• Identifying methods
• Industry development
• Industry life cycles
• Information entropy
• Radial basis function neural networks
• RBF Neural Network

In recent years, research about industry development stage identification is taken seriously increasingly and great achievements have been made. Against the background of urban economy, an identifying method of city's industry development stage is put forward based on integration of Rough Sets (RS) and Artificial Neural Network (ANN). At first, the continuous attribute values are discretized using fuzzy clustering algorithm based on Maximum Discernibility Value (MDV) search method and information entropy. And then the major attributes are reduced by rough sets. At last, the Radial Basis RBF) neural network is trained with training samples and the industry life cycle stages of testing samples are identified. The analysis results taking 669 industries of Dalian city as samples show that the fuzzy clustering algorithm based on MDV and information entropy can improve the discretization performance effectively. Compared with normal fuzzy evaluation method, the predicting precision of integration method is higher and it is an efficient and practical tool to identify development stage of city's industry.

• 出版日期2013
• 单位中国矿业大学（北京）