基于改进BP算法的入侵检测神经网络方法

论文利用基于Cauchy误差估计器的BP算法对入侵检测系统进行了研究,改进的BP算法较传统BP算法具有收敛速度快、正确检测率高的优点,实验结果表明:该方法较为圆满地解决了入侵检测系统所需要的实时性、适应性、可用性、可靠性和准确性等方面的要求,这为入侵检测提供了一个新的研究手段与方法。

The Neural Network Method for Intrusion Detection Based on Improved BP Algorithm

This paper makes use of BP algorithm which is based on error estimator of Cauchy to study the IDS.Compared with traditional BP algorithm,the new BP algorithm it brings up has higher speed in constringency and is more precise in detection.The results of the experiment indicate that the methods solve the need of real -time ,adaptability,availability,dependability and veracity for the IDS' products with satisfactoriness.Thus,it provides a new research method for intrusion detection.