CAJ | 학술논문

针对网络入侵检测，提出一种基于小波核主成分分析和差分进化极限学习机相结合的方法。首先采用核主成分分析法对原始数据进行非线性降维处理，为了进一步提高核PCA的非线性映射能力，引用小波核函数作为核PCA的核函数。然后采用极限学习机对处理后的数据进行分类识别，针对初始权值随机选择造成极限学习机性能不稳定的问题，采用差分进化算法来获得最优的初始权值。实验结果表明该算法可以有效提高入侵检测的识别率，降低误报率和漏报率。
침대망락입침검측，제출일충기우소파핵주성분분석화차분진화겁한학습궤상결합적방법。수선채용핵주성분분석법대원시수거진행비선성강유처리，위료진일보제고핵PCA적비선성영사능력，인용소파핵함수작위핵PCA적핵함수。연후채용겁한학습궤대처리후적수거진행분류식별，침대초시권치수궤선택조성겁한학습궤성능불은정적문제，채용차분진화산법래획득최우적초시권치。실험결과표명해산법가이유효제고입침검측적식별솔，강저오보솔화루보솔。
For network intrusion detection,we propose such a method which combines the wavelet kernel PCA and DE optimised extreme learning machine.First,the kernel principal component analysis (PCA)is applied to conduct the nonlinear dimensionality reduction on original data,in order to further improve nonlinear mapping ability of kernel PCA,wavelet kernel function is introduced as its kernel function. Then the extreme learning machine is used for the classification and recognition of the processed data,and the differential evolution (DE) algorithm is used to obtain the optimal initial weights for the unstable performance of the extreme learning machine caused by random selection of initial weights.Experimental results show that the algorithm proposed can effectively improve the recognition rate of intrusion detection and reduce the rates of false positives and false negatives.