基于主元神经网络和SVM的入侵特征抽取和检测

Principal Component Neural Networks Feature Extraction with SVM for Anomaly Detection

目前基于机器学习的入侵检测研究都是从提高检测精度的分类器算法设计出发,大多未考虑对样本特征的分析。文章提出了一种基于特征抽取的异常检测方法,应用主元神经网络(PCNN)抽取入侵特征,再应用SVM检测入侵。采用广义Hebb学习规则训练线性主元神经网络,SVM采用基于网格粒度搜索获得最优参数。利用KDD99数据集,将线性PCNN-SVM与SVM进行比较,结果显示在不降低分类器性能的情况下,PCNN特征抽取方法能对输入数据有效降维。

Very little work on feature extraction has been taken in the field of network anomaly detection.This paper proposes the application of principal component neural networks for intrusion feature extraction.The extracted features are employed by SVM for classification.The MIT's KDD Cup 99 dataset is used to evaluate the proposed method compared to SVM without application of feature extraction,which clearly demonstrates that PCNN-based feature extraction method can greatly reduce the dimension of input space without degrading or even boosting the classifiers' performance.