【摘要】 针对神经网络的结构特性,提出将层次分析法(AHP)与BP神经网络相结合的思想,即利用层次分析法在筛选对因变量最有影响的因素(自变量)方面独特的性质,来决定BP神经网络的输入变量的个数。并在此思想上建立了AHP与BP相结合的网络模型,又设计了相应的算法。该模型不仅解决了输入待定的神经网络输入维数难确定的问题,而且加快了网络的收敛速度,增强了网络的适应能力。并将新建立的模型应用到经济预测中,得出了比常规经济学模型更优的效果。更多还原

【Abstract】 Aiming at the characteristics of neural network structure, an idea of connecting AHP and BP is presented. The idea is to use the character that AHP can select the independent variables that affect the dependent variables essentially, and to decide the inputlay units of BP neural network. Moreover, this idea is adopted to build the model that connects AHP and BP, and to design the algorithm. This model can make the selection of inputlay units easily, and improve the rate of studying of the neural network. Then the method is used in the economic fields. The conclusion shows that this method is superior to the statistical modeling approach. 更多还原