基于BP网络的入侵检测方法

A Back-propagation Neural Network-based Intrusion Detection Ap proach

论文提出了一种基于BP神经网络的入侵检测方法。该方法对特征数据进行了预处理,利用改进的BP算法的学习能力和快速识别能力,实现了对用户行为的检测,尤其是在识别以前没有观察到的未知攻击方面具有较好的性能。

This paper describes new back-propagation neural network-based intrusion detection approach.Based on the feature data that should be pretreated,the approach employs improving back-propagation arithmetic that provide the ca-pability of learning and quick classification,and can be used for detection of user behavior.In particular,the approach gains good performance in recognizing future unseen attacks.