Intrusion Detection based on KELM with Levenberg-Marquardt optimization

Intrusion is an illegitimate event that can either be active or passive in a network. In this work, we propose an Intrusion Detection System (IDS) on the basis of Kernel Extreme Learning Machine (KELM) clubbed with Levenberg-Marquardt optimization technique. We incorporate KELM in this work, because of its efficiency in pattern recognition. Levenberg-Marquardt optimization technique is employed because of its efficiency over other gradient descent techniques. The proposed system is compared with several existing works and the results obtained are satisfactory.