A study in using neural networks for anomaly and misuse detection

摘要: Current intrusion detection systems lack the ability to generalize from previously observed attacks detect even slight variations of known attacks. This paper describes new process-based approaches that provide behavior recognize future unseen behavior. The approach employs artificial neural networks (ANNs), and can be used for both anomaly in order novel misuse These techniques were applied a large corpus data collected by Lincoln Labs at MIT an system evaluation sponsored U.S. Defense Advanced Research Projects Agency (DARPA). Results applying these against DARPA are presented.