提出了基于粗糙集理论和支持向量机(SVM)的入侵检测方法,利用粗糙集约简算法对样本集进行特征约简,删除对入侵检测结果影响不大的冗余特征,从而有效地降低了样本集的维数,解决了SVM训练时间长,样本集占用的存储空间过大的问题.实验证明,该方法能在不影响SVM检测精度的情况下,缩短SVM的训练和检测时间,有效地提高SVM的检测效率.
Proposed a method of detecting intrusion using SVM based on Rough Ret Theory(RST). The reduction algorithm based on RST is used to eliminate the redundant features of sample dataset, in order to reduce the space dimension of the sampie data. Using this method, it can overcome the shortages of SVM - time-consuming of training and massive dataset storage. Experiment results show , using this method can improve the efficient of SVM by reduce the training time and test time ,but not reduce the accuracy of SVM.