Research on modeling method of artificial neural network based on DEA

Abstract

A modeling method of artificial neural network (ANN) is proposed. Experimental data were evaluated and projected by using data envelopment analysis (DEA), a widely used method to evaluate relative efficiency between decision making units. The data would become more scientific and reasonable, and all of them could be used for the modeling of ANN. Example shows that the model of ANN, which is gained by training these data, is DEA effective. Hence, it is a new method for optimal data utilizing and modeling. This method is useful to the research, which may only get limited and high cost data after several times or several years of experiments.