基于神经网络的检测器生成算法研究与应用

Research and Applications on Detector Generation Algorithm Based on Neural Networks

否定选择算法是用于生成人工免疫系统检测器的重要算法,而有效的检测器生成算法是入侵检测的核心问题。文章针对现有检测器生成算法存在自适应差的问题,通过对人工免疫系统中否定选择算法原理的分析,提出了一种基于神经网络的检测器生成算法。该算法利用高效的神经网络训练,使得检测器生成算法具有很好的动态适应能力。实验结果表明,该算法应用于入侵检测,提高了检测率,降低了误检率,整体检测性能较好。

Negative selection algorithm （NSA） is an important method of generating artiifcial immune detectors, and efficient detector generation algorithm is the kernel of intrusion detection. Aiming at conventional NSA detectors are not adaptive for dealing with time-varying circumstances, this paper analyzed the negative selection algorithm principle in an artiifcial immune system, and put forward a detector generation algorithm based on neural networks. Taking advantage of efifcient neural networks training, it has the distinguishing capability of adaptation. Experimental results show that the algorithm performs well that it improves the detection rate and reduces the false dtection rate.