Agents and Neural Networks for Intrusion Detection

摘要

Up to now, several Artificial Intelligence (AI) techniques and paradigms have beensuccessfully applied to the field of Intrusion Detection in Computer Networks. Most of themwere proposed to work in isolation. On the contrary, the new approach of hybrid artificial intel- intelligentsystems, which is based on the combination of AI techniques and paradigms, is probingligent to successfully address complex problems. In keeping with this idea, we propose a hybrid useof three widely probed paradigms of computational intelligence, namely Multi-Agent Systems,Case Based Reasoning and Neural Networks for Intrusion Detection. Some neural modelsbased on different statistics (such as the distance, the variance, the kurtosis or the skewness)have been tested to detect anomalies in packet-based network traffic. The projection method ofCurvilinear Component Analysis has been applied for the first time in this study to performpacket-based intrusion detection. The proposed framework has been probed through anomaloussituations related to the Simple Network Management Protocol and normal traffic.