A new intrusion detection method based on data-oriented classification of attacks

The most acute problem for misuse detection method is its inability to detect new kinds of attacks. A new detection method based on data-oriented classification of attacks is proposed to solve this problem. After analyzing its significance, a practical scheme which uses relevant feature subset codes clustering is designed. Applying Concept Hierarchy Generation for attack Labels (CHGL), inductive learning algorithms can learn attack profiles on high concept levels. Experimental results show the advantage of this method.