Class imbalance problem in the Network Intrusion Detection Systems
Sireesha Rodda, Uma Shankar Rao Erothi · 2016
Network Intrusion Detection Systems must effectively identify security threats and protect the applications. The focus of the paper is the presence of class imbalance problem in intrusion datasets. An efficient intrusion detection system must accurately identify all threats even if they form a small fraction of the intrusion data. The effect of class imbalance on the benchmark NSL_KDD dataset is evaluated using four popular classification techniques and the results are analyzed.