摘要
为了提高入侵检测系统的准确率,提出一种基于乌鸦搜索算法的反向传播(CSA-BP)神经网络模型。BP神经网络是解决非线性问题的重要方法,但是其预测能力容易受到初始参数的影响。针对这一问题,将相对百分误差作为模型的目标函数,通过乌鸦搜索算法极强的全局搜索能力找到最优权值和阈值。然后,利用5组标准的数据集对CSA-BP模型进行验证。最后,将CSA-BP算法用于入侵检测系统,结果表明,该算法使入侵检测系统准确率更高,达到了96.6%,且加快了收敛速度。
In order to improve the accuracy of the intrusion detection system,a back propagation neural network model based on the crow search algorithm(CSA-BP)is proposed.BP neural network is an important method to solve nonlinear problems,but its predictive ability is easily affected by the initial parameters.To solve this problem,the relative percentage error is used as the objective function of the model,and the optimal weight and threshold are found through the strong global search ability of the crow search algorithm.Then,the CSA-BP model is validated with five standard datasets.Finally,the CSA-BP algorithm is used in the intrusion detection system.The results show that the proposed algorithm makes the intrusion detection system more accurate,reaching 96.6%,and speeds up the convergence.
作者
蓝吕盈
唐向红
顾鑫
陆见光
Lan Lüying;Tang Xianghong;Gu Xin;Lu Jianguang(Key Laboratory of Advanced Manufacturing Technology,Ministry of Education,Guizhou University,Guiyang,Guizhou 550025,China;School of Mechanical Engineering,Guizhou University,Guiyang,Guizhou 550025,China;Stata Key Laboratory of Public Big Data,Guizhou University,Guiyang,Guizhou 550025,China)
出处
《激光与光电子学进展》
CSCD
北大核心
2021年第6期148-155,共8页
Laser & Optoelectronics Progress
基金
贵州省公共大数据重点实验室开放基金资助项目(2017BDKFJJ019)
贵州省留学回国人员科技活动择优资助项目-优秀类项目(2018.0002)。
关键词
图像处理
入侵检测
反向传播神经网络
乌鸦搜索算法
参数优化
image processing
intrusion detection
back propagation neural network
crow search algorithm
parameter optimization
作者简介
唐向红,E-mail:lanlym249@163.com。
