Research On Neighborhood Detector Generating Algorithm And Model Based On Clustering

With the rapid development of network technology,more and more people began to focus on intrusion detection immunity. The immune principle, characteristics, system structure and the related algorithms are combined with intrusion detection,for example the legitimate user, legal authorization behavior as self samples. Viruses, hackers and other actions are non-self samples.So naturally immune technology is applied in intrusion detection system naturally.In recent years,as an efficient intrusion,immune intrusion detection technology has become the focus of modern network security research.The essay could introduce the present situation of immune intrusion research at home and abroad,and analysis the existing problems and introduce a new space of immune intrusion-Neighborhood shape space.The neighborhood negative selection algorithm is used to generate the training detector,To address this issue,this paper makes in-depth study on NNS and clustering method and combines them together,this paper proposes a novel Neighborhood Detector Algorithm Based on Clustering(NDAC). The self-samples are mapped to neighborhood space and they are used to cluster, Meanwhile, random detectors are trained and become mature neighborhood detectors. The algorithm generates detectors by shortening the time and solving the high overlap problem.There are many problems, for example defects in shape space,Poor adaptability, model running rate slowly and high overlap. This paper designs the neighborhood detector module, neighborhood memory module and self samples module based on the neighborhood immune intrusion detection model.At the same time, this model can adapt to the environment better, generate rapidly, test detection rate higher.The experimental results show that neighborhood detector algorithm based on clustering can reduce the amount of computation, save time, and improve efficiency. At the same time, this model has good adaptability and high rate of detection.