Attack Detection Based on Data Mining Techniques

In light of the security challenges posed by the reality of today, where the Internet and exchange information are an integral part of our daily lives, we live in a world where data requirements have become dynamic, where things are permanently changing. In order to provide security and decrease the damage of information system caused by attacks on the network, it is important to provide it with Intrusion Detection system (IDS). In this paper, we present intrusion detection model based on Feature extraction and two-stage classifier module, designed to detect anomaly activities. The proposed model using Principal Component Analysis (PCA) of Feature extraction to map the high dimensional dataset to a lower one with effective features. We then apply a two-stage classification module utilizing Nave Bayes and C4.5 to identify abnormal behaviors. The experiment results using NSL-KDD dataset shows that Our model outperforms the previous model for detection low-frequency attacks.