论文部分内容阅读
本文应用Bayesian正则化算法改进BP神经网络泛化能力。通过对湖北省1985年—2004年关于经济发展水平的数据进行拟和以及预测,结果表明采用Bayesian正则化算法比相同条件下采用其他改进算法泛化能力要好,收敛速度快、预测精度高,方法简单,操作方便。实例分析表明,贝叶斯正则化算法优化BP神经网络的方法是令人满意的,对经济预测有良好的预测效果。
In this paper, Bayesian regularization algorithm is used to improve the generalization ability of BP neural network. Through the data of Hubei Province from 1985 to 2004 on the level of economic development and forecasting, the results show that the use of Bayesian regularization algorithm under the same conditions using other improved algorithm generalization ability is better, the convergence speed, the prediction accuracy is high, the method Simple, easy to operate. The case study shows that Bayesian regularization algorithm is a satisfactory method to optimize BP neural network and has a good predictive effect on economic forecasting.