Approximation capabilities of multilayer feedforward networks
Neural Networks
A framework for constructing features and models for intrusion detection systems
ACM Transactions on Information and System Security (TISSEC)
Layered Approach Using Conditional Random Fields for Intrusion Detection
IEEE Transactions on Dependable and Secure Computing
Intrusion detection system based on multi-class SVM
RSFDGrC'05 Proceedings of the 10th international conference on Rough Sets, Fuzzy Sets, Data Mining, and Granular Computing - Volume Part II
| Hi-index | 0.00 |
Today, with more and more computers getting connected to public accessible networks like Internet, computer systems are more and more susceptible to attacks. There is a need of effective intrusion detection systems (IDS) to protect computers from these unauthorized or malicious actions. Different soft-computing based methods have been proposed in recent years for the development of intrusion detection systems. In this paper we are proposing a contemporary approach for network intrusion detection which is significantly improvises the prediction of network intrusions. The proposed technique is a fusion of efficient data mining techniques such as Fuzzy C-means (FCM) clustering, neural network (NN) and Support Vector Machine (SVM). The experiments and evaluations of proposed method were performed with Knowledge Discovery and Data mining (KDD) Cup 99 intrusion detection dataset and we have used sensitivity, specificity, accuracy, precision and F-value as the evaluation metric parameters. Our approach achieved detection accuracy of about 99.96% for DOS attacks, 99.73% for PROBE, 99.93% for R2L and 99.87% for U2R attacks. Our proposed technique yielded very good results and was compared with the other existing techniques to prove its validity.