摘要
提出一种基于BP神经网络的异常入侵检测方法,由于BP神经网络是一种基于误差反向传播算法的多层前馈神经网络,具有对不确定性的学习与适应能力,可以很好的满足入侵检测分类识别的需求.对“KDD Cup 1999 Data”网络连接数据集进行特征选择和标准化处理之后用于训练神经网络并仿真实验,得到了较高的检测率和较低的误报率.仿真实验表明,基于BP神经网络的入侵检测方法是有效的.
An anomaly intrusion detection approach using BPN is proposed in this paper. BPN is a multilayer Feedforward Neural Network based on Error Back-propagation algorithm. Being able to adapt itself to and fault-tolerant to imprecise data and uncertain information it seems to be an appropriate approach to intrusion detection. Simulated experiments with KDD Cup 1999 network traffic connections which have been preprocessed with methods of features selection and normalization have shown that BPN model is effective for intrusion detection owing to excellent performance of the higher attack detection rate with lower false positive rate.
出处
《南京晓庄学院学报》
2006年第6期82-86,共5页
Journal of Nanjing Xiaozhuang University
关键词
入侵检测
异常检测
神经网络
BP算法
intrusion detection
anomaly detection
neural network
BP algorithm
