Neural Network Construction of Complex System Based on Rough Set Theory
ZOU Gu-shan, CAI Yan-guang, LUO Shi-liang (College of Automation Engineering, Guangdong University of Technology, Guangzhou 510090, China)
An NN model of complex system based on rough set theory is designed, and the algorithm of this model is described in detail.It is applied some natures of the discernable matrix and rough set theory for reduction of the basic knowledge performance of the complex system with plentiful data, the multitudinous dimension performance criteria and the multi-decision-making criteria effectively. And then a less complicated neural network is constructed. A method to solve the construction problem of complex system NN model is presented.Finally, the effectiveness of the result obtained is demonstrated by an example.