Abstract: In the era of internet, the network security has become the key foundation for many web applications. Intrusion detection is one which resolves this kind of a problem of internet security. Improperness of intrusion detection system (IDS) has given an opportunity for data mining technique to make several contributions to the field of intrusion detection. In the recent years, many researchers are using data mining technique for building IDS. Here, we propose a new approach by utilizing the data mining techniques such as neuro-fuzzy logic and support vector machine(SVM) helping IDS to attain high detection rate. The proposed technique has four major steps: primarily, k-means clustering is used to generate different training subsets. Then, based on obtained trained data subset, different neuro-fuzzy models are trained. Subsequently, a vector for SVM is formed and in the end, classification using radical SVM is to detect intrusion has happened or not. Experimental results show that out proposed approach do better than BPNN, multiclass SVM and other well-known major methods such as decision tree and Columbia model in terms of sensitivity, specificity and in particular detection accuracy.
Keywords: Intrusion detection system, Fuzzy-Neural networks, Support Vector Machine (SVM), K-means clustering.