利用BP神经网络组合能够较好地模拟在各种不确定因素影响下因果变量之间的内在关系。建立了基于人工神经网络的卫生总费用预测模型,该模型的网络结构由输入层(1个节点)、隐层(7个节点)和输出层(1个节点)组成。采用改进的BP算法对7组学习样本进行训练,得到各节点间的连接权和阈值,然后用优化好的网络进行卫生总费用预测。预测结果表明,利用该方法建立的模型预测误差在1%以内,十分精确。 Using BP neural network combination can better simulate the internal relationship between causal variables under the influence of various uncertainties. A total health cost prediction model based on artificial neural network was established. The network structure of the model consists of input layer (1 node), hidden layer (7 nodes) and output layer (1 node). The improved BP algorithm was used to train 7 groups of learning samples, and the connection weights and thresholds between nodes were obtained. Then the optimized total network cost forecast was made using the optimized network. The prediction results show that the model prediction error established by this method is within 1%, which is very accurate.