Mining Network Stream Data for Self Learning Networks

Detecting attacks on networks is a challenging task. Classifying packets as either normal packets or malicious packets is important to detect any attacks. This paper discusses methods to classify network traffic data as benign versus malicious. The original NSL-KDD dataset is classified using Artificial Neural Network. A taxonomy and survey of artificial neural network classification system is presented based on current and previous work. The effectiveness of the proposed algorithm is compared with Tree Based Classifier, logistic regression, support vector machine and the validity of the proposed algorithm is verified by using MATLAB.