论文部分内容阅读
BP神经网络能较好处理非线性化数据,但传统BP神经网络存在着局限性,为了提高神经网络运算的精确度,通过权值和学习率共同优化,并采用贝叶斯正则化算法训练神经网络,形成了基于改进型BP神经网络的管理信息系统开发风险评价模型,经测算,该模型输出值与实际值高度吻合,模型可接受度较高,并且与传统BP神经网络相比,改进型BP神经网络的相对误差更小。
BP neural network can better deal with the nonlinear data, but the traditional BP neural network has its limitations. In order to improve the accuracy of the neural network operation, the weights and learning rates are jointly optimized, and the Bayesian regularization algorithm is used to train the neural Network, a risk assessment model of management information system based on improved BP neural network has been formed. The output value of this model is highly consistent with the actual value and the model is more acceptable. Compared with the traditional BP neural network, The relative error of BP neural network is smaller.