CAJ | 학술논문

为了保证网络的安全,针对极限学习机在网络入侵检测过程中参数优化的难题,提出一种改进布谷鸟搜索算法优化极限学习机的网络入侵检测模型。首先将极限学习机参数编码为布谷鸟巢位置,并以网络入侵检测正确率作为ELM参数优化目标,然后通过模拟布谷鸟繁育行为找到极限学习机的最优参数,建立网络入侵检测分类器,最后在Matlab 2012平台上采用KDD99数据集进行仿真实验。结果表明,MCS-ELM提高了网络入侵检测正确率,可以满足网络入侵检测在线检测要求。
위료보증망락적안전,침대겁한학습궤재망락입침검측과정중삼수우화적난제,제출일충개진포곡조수색산법우화겁한학습궤적망락입침검측모형。수선장겁한학습궤삼수편마위포곡조소위치,병이망락입침검측정학솔작위ELM삼수우화목표,연후통과모의포곡조번육행위조도겁한학습궤적최우삼수,건립망락입침검측분류기,최후재Matlab 2012평태상채용KDD99수거집진행방진실험。결과표명,MCS-ELM제고료망락입침검측정학솔,가이만족망락입침검측재선검측요구。
In order to guarantee network security, in view of the parameter optimization problem of e extreme learning machine in the process of network intrusion detection, a network intrusion detection model is proposed based on modified cuckoo search algorithm optimizing extreme learning machine for network intrusion detection system. First-ly, the extreme learning machine parameters are encoded as the cuckoo’ s nest location and the network intrusion de-tection correct rate is taken as the optimization goal of ELM parameters, and then he optimal parameters of the extreme learning machine is found by simulating the cuckoo bird breeding behavior and establishing the network intrusion detec-tion classifier, finally, the simulation experiments are carried out on Matlab 2012 platform by using KDD99 data set. The results show that the proposed model can improve the correct rate of network intrusion detection and can meet on-line detection requirements of network intrusion detection.