CAJ | 학술논문

为了有效地提高入侵检测系统的检测率并降低误报率，提出采用属性约简方法对高维入侵检测数据进行特征选择，剔除无关的属性输入来提高检测效果，将混沌免疫遗传算法引入神经网络学习过程用以进行入侵检测，与传统BP神经网络检测结果进行比较，实验结果表明，将该方法用于入侵检测是切实可行的。
위료유효지제고입침검측계통적검측솔병강저오보솔，제출채용속성약간방법대고유입침검측수거진행특정선택，척제무관적속성수입래제고검측효과，장혼돈면역유전산법인입신경망락학습과정용이진행입침검측，여전통BP신경망락검측결과진행비교，실험결과표명，장해방법용우입침검측시절실가행적。
In order to effectively improve the detection rate of intrusion detection system and reduce the false alarm rate, the method of attribute reduction of high-dimensional data in intrusion detection feature selection is proposed. Attribute input irrelevant is weeded out to improve the detection effect. The chaos immune genetic algorithm is used in neural net-work learning process for intrusion detection. Compared with the traditional BP neural network detection results, the experi-mental results show that the method used in intrusion detection is feasible.