BP神经网络在入侵检测中的应用

针对传统入侵检测系统在误报率、漏报率上的缺陷，把神经网络应用于专业入侵检测系统中，利用神经网络的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，最后给出了该算法的理论推导和算法的具体实现步骤。与传统的入侵检测算法相比，该方法的最小平均误差比传统算法的最小平均误差小，且学习速度快。实验结果也说明了该方法的快速性、有效性、稳定性。

The Application of BP-Neural Network in the Intrusion Detection

In allusion to the insufficiencies such as missing report rate and false alarm rate, the paper applys neural network into intrusion detection, achieves a goal of user behavior with BP algorithm's learning ability and fast recognition ability. At last the paper gives out a theoretical derivation and the steps of algorithm. Compared with old algorithm, new algorithm's minimum average error is smaller than the old algorithm's, furthermore, its learning speed is faster. From the results of experiment, we can se...