A New Intrusion Detection System Using Class and Sample Weighted C-support Vector Machine

whenever an intrusion occurs, the security of a computer system is compromised. Presently there are a lot of algorithms applied in intrusion detection systems. The SVM is one of the most successful ones in the data mining area, but its biasing behavior with uneven datasets limits its use. Shu-Xin Du proposed an improved approach named weighted SVM to solve this problem. However, Weighted SVM considers different penalty parameters about class only and ignores importance among different samples. In this paper, we introduced class and sample weighted factors respectively and propose a new method, namely, Class and Sample Weighted C-Support Vector Machine (CSWC-SVM) to solve the problem. Furthermore we construct a decision model. Experimental simulations with KDD Cup 1999 Data proved our approach works well and outperforms other approaches such as the standard C-SVM and Weighted SVM in terms of accuracy, false positive rate, and false negative rate.