Combining the neural network, support vector machine and KNN in intrusion detection system

Immunity of computer networks has a critical role in computer systems. The prominent growth of the application of computer networks resulted the abuse of some intruders. Therefore, in order to protect the system against the risks, a number of software methods have been provided under the caption of intrusion detection system. The purpose of intrusion detection is to detect the illegal intrusions, destruction and damaging the computer systems and networks by bath internal users and intruder’s. In this article, a combined of MLP neural networks and SVM algorithms and KNN are used in order to design the portal of intrusion detection system. The proposed method has been tested both KDDcup99 dataset. The results comparing with the results of the other methods showed a very optimal rate in the precision and the correctness in intrusion detection.

کلیدواژه ها:

Intrusion detection system ، SVM ، KNN ، MLP neural networks