Detecting Denial of Service Attack Using Principal Component Analysis with Random Forest Classifier

--Nowadays, computer network systems plays gradually an important role in our society and economy. It became a targets of a wide array of malicious attacks that invariably turn into actual intrusions. This is the reason that computer security has become an essential concern for network administrators. In this paper, an exploration of anomaly detection method has been presented. The proposed system uses a statistical method called principal component analysis to filter the attributes and random forest classifier is used to detect various attack present in Denial of Service using NSL-KDD dataset. The principal component Analysis filters attributes drastically to improve classification performance. Regarding to the task of intrusion detection a new method of random forest classifier is used to improve accuracy. Experimental result shows that the proposed method can achieve high detection rate than other existing machine learning techniques. This approach is dynamic in the sense that the model is updated based on the variations of its input. Our experiments revealed relevant results that can effectively be used to classify Denial of Service attacks. [email protected] Keyword--Intrusion Detection, Principal component analysis, Random Forest, NSL-KDD dataset