An additive decision rules classifier for network intrusion detection

Abstract

This paper presents an additive decision rules binary classifier applied for network intrusion detection. The classifier is optimized by a multi objective evolutionary algorithm in order to maximize both the classification accuracy and the coverage level (percentage of items that are classified, in opposite to items unable to be classified). Preliminary results provides very good accuracy in detecting attacks which make this relatively simple classifier very suitable to be applied in the studied domain. © 2011 Springer-Verlag.