CAJ | 학술논문

为了提高网络入侵检测率，提出一种反向学习粒子群算法和多层次分类器相融合的网络入侵检测模型。首先将反向学习粒子群算法优化最小二乘支持向量机，以提高分类性能；然后利用由粗到精策略构造多层的网络入侵分类器降低计算时间杂度复；最后采用KDD 99数据集进行仿真测试。仿真结果表明，相对于其他检测模型，该模型不仅提高了网络入侵检测率，降低了入侵检测误报率，同时加快了入侵检测速度，为网络安全提供了有效保证。
위료제고망락입침검측솔，제출일충반향학습입자군산법화다층차분류기상융합적망락입침검측모형。수선장반향학습입자군산법우화최소이승지지향량궤，이제고분류성능；연후이용유조도정책략구조다층적망락입침분류기강저계산시간잡도복；최후채용KDD 99수거집진행방진측시。방진결과표명，상대우기타검측모형，해모형불부제고료망락입침검측솔，강저료입침검측오보솔，동시가쾌료입침검측속도，위망락안전제공료유효보증。
In order to improve detection rate of the network intrusion,we propose a network intrusion detection model with the fusion of reverse learning particle swarm optimisation (RLPSO)and multilayer classifier.First,the RLPSO is employed to optimise LSSVM to improve the classification performance;then the coarse-to-fine strategy is used to construct multi-layer network intrusion classifier to reduce the complexity of computation time;finally,the simulation test is carried out using KDD 99 data.Simulation results show that compared with other detection models,the proposed model improves network intrusion detection rate and lowers the false positive rate of intrusion,meanwhile it also accelerates the intrusion detection speed,and provides effective guarantee for network security.