摘要
为了提高网络流量的预测精度,克服BP神经网络预测过程中存在收敛速度慢、易陷入局部极小值的缺点,提出改进BP神经网络的网络流量预测模型。该模型引入动量因子和自适应学习速率来改进BP神经网络。仿真结果表明,改进BP神经网络预测的结果误差更小,精确度更高。
In order to improve the prediction accuracy of network traffic and overcome the shortcomings of slow convergence speed and easy to fall into local minimum in the process of BP neural network prediction,an improved BP neural network traffic prediction model is proposed.This model introduces momentum factor and adaptive learning rate to improve BP neural network.The simulation results show that the improved BP neural network has smaller prediction error and higher accuracy.
作者
吉珊珊
柯钢
JI Shanshan;KE Gang(Department of Computer Engineering,Dongguan Polytechnic,Dongguan 523808)
出处
《计算机与数字工程》
2020年第7期1682-1686,共5页
Computer & Digital Engineering
基金
东莞职业技术学院示范建设专项资金项目(编号:政201819)
2017广东省教育厅青年创新人才类项目(编号:2017GkQNCX116)资助。