计算机工程与应用
計算機工程與應用
계산궤공정여응용
COMPUTER ENGINEERING AND APPLICATIONS
2014年
24期
86-90,95
,共6页
偏最小二乘%特征提取%核心向量机%异常入侵检测%支持向量机
偏最小二乘%特徵提取%覈心嚮量機%異常入侵檢測%支持嚮量機
편최소이승%특정제취%핵심향량궤%이상입침검측%지지향량궤
Partial Least Square(PLS)%feature extraction%Core Vector Machine(CVM)%anomaly intrusion detection%Support Vector Machine(SVM)
为提高异常入侵检测的效率,提出一种混合偏最小二乘特征提取和核心向量机算法的入侵检测模型。模型使用偏最小二乘算法在入侵数据集上进行主成分提取,在此基础上构建特征集,引入适用于解决大规模样本训练问题的核心向量机算法,在特征集上建立入侵检测模型,使用该模型对异常入侵行为进行检测和判断。通过基于KDD99数据集上的入侵检测实验,验证了混合模型的可行性和有效性。
為提高異常入侵檢測的效率,提齣一種混閤偏最小二乘特徵提取和覈心嚮量機算法的入侵檢測模型。模型使用偏最小二乘算法在入侵數據集上進行主成分提取,在此基礎上構建特徵集,引入適用于解決大規模樣本訓練問題的覈心嚮量機算法,在特徵集上建立入侵檢測模型,使用該模型對異常入侵行為進行檢測和判斷。通過基于KDD99數據集上的入侵檢測實驗,驗證瞭混閤模型的可行性和有效性。
위제고이상입침검측적효솔,제출일충혼합편최소이승특정제취화핵심향량궤산법적입침검측모형。모형사용편최소이승산법재입침수거집상진행주성분제취,재차기출상구건특정집,인입괄용우해결대규모양본훈련문제적핵심향량궤산법,재특정집상건립입침검측모형,사용해모형대이상입침행위진행검측화판단。통과기우KDD99수거집상적입침검측실험,험증료혼합모형적가행성화유효성。
To improve the efficiency of anomaly detecting intrusions, a hybrid model is proposed based on Partial Least Square(PLS)feature extraction and Core Vector Machine(CVM)algorithms. Principal elements are extracted from the intrusion data set by the feature extraction of PLS algorithm to establish the feature set, and then the anomaly intrusion detec-tion model for the feature set is constructed by virtue of speediness superiority of CVM algorithm in processing large-scale sample data. Anomaly intrusion actions are checked and judged using this model. Intrusion detection experiments based on KDD99 data set verify the feasibility and validity of the hybrid model.