把神经网络理论应用到投入产出分析中，由此建立起一种新的投入－产出模型，把这一模型中的某些变量理想化并进行变换后即可得到列昂节夫（Ｌｅｏｎｔｉｅｆ）模型。为此，本文把基于神经网络的新模型与列昂节夫模型进行了多方面的比较，并讨论了新模型系统的Ｌｙａｐｕｎｏｖ函数及关联强度的变化规则等 The neural network theory is applied to the input-output analysis, thus establishing a new input-output model, some of the variables in this model idealized and transformed to get Leontief (Leontief) )model. In this paper, we compare the new model based on neural network with the Leonov model and discuss the Lyapunov function of the new model system and the change rules of the correlation strength