摘要
提出一种改进的蚁群算法并与传统的BP神经网络相结合用于入侵检测,它既克服了BP传统神经网络的权值确定难度较大、收敛速度慢易陷入局部最小等缺陷,也通过BP神经网络的梯度信息弥补了单独使用蚁群算法所面临的不足.仿真实验结果表明,与传统方法相比,本方法步骤简化,速度及测试精度明显提高.
This paper proposes an improved ant colony algorithm with the traditional BP network com- bination used in intrusion detection, it has overcome the traditional BP network weights to determine the difficulty of slow convergence speed, easy to fall into local minimum and other defects, but also by the BP network gradient information for individual use ant colony algorithm the deficiency. The simula- tion results show that compared with the traditional method, the method steps are simplified, speed and accuracy are increased obviously.
出处
《福州大学学报（自然科学版）》
CAS
CSCD
北大核心
2013年第5期845-849,共5页
Journal of Fuzhou University(Natural Science Edition)
关键词
蚁群算法
神经网络
入侵检测
ant colony algorithm
neural network
intrusion detection
