Online Boosting Based Intrusion Detection in Changing Environments

Intrusion detection is an active research field in the development of reliable web-based information systems, where many artificial intelligence techniques are exploited to fit the specific application. Although some detection algorithms have been developed, they lack the adaptability to the frequently changing network environments, since they are mostly trained in batch mode. In this paper, we propose an online boosting based intrusion detection method, which has the ability of efficient online learning of new network intrusions. The detection can be performed in real-time with high detection accuracy. Experimental results show the advantage of the method in the intrusion detection application. Categories and Subject Descriptors K.6.5 [Management of Computing and Information Systems]: Security and Protection--Invasive software (e.g., viruses, worms, Trojan horses), Unauthorized access (e.g., hacking, phreaking); C.2.3 [Computer-Communication Networks]: Network Operations--Network mon...

• Type: Conference
• Year: 2008
• Where: ICAI
• Authors: Yanguo Wang, Weiming Hu, Xiaoqin Zhang
• Keywords: Artificial Intelligence, ICAI 2008, Intrusion Detection, Intrusion Detection Application, Intrusion Detection Method