摘要： 提出一种改进的BP算法，该算法中学习速率 不是凭经验给出的固定值，而是在学习过程中计算出的最优值αk。实验结果表明，与传统的BP算法相比，该算法可以减少学习时间，提高学习速度和网络入侵检测系统(NIDS)的检测率，可以解决当前NIDS普遍存在的虚警率过高问题。

关键词: 网络入侵检测系统, 神经网络, BP算法

Abstract: An improved BP algorithm is proposed,in the algorithm,the learning rate α is not given by experts, but the best value is computed during learning. The experimental results indicate that this improved algorithm can increase learning speed and shorten training time. have become a part of security infrastructures. As high false alarm rate exists in current network intrusion detection systems, this improved algorithm can improve the intrusion detection rate.

Key words: network intrusion detection systems, neural network, BP algorithm

中图分类号: 

• TP393.08