FUSION OF RECIRCULATION NEURAL NETWORKS FOR REAL-TIME NETWORK INTRUSION DETECTION AND RECOGNITION

Intrusion detection system is one of the essential security tools of modern information systems. Continuous development of new types of attacks re quires the development of intelligent approaches for intrusion detection capable to detect newest attacks. We present recirculation neural network based approach which lets to detect previously unseen attack types in real-time mode and to further correct recognition of this types. In this paper we use recirculation neural networks as an anomaly detector as well as a misuse detector, ensemble of anomaly and misuse detectors, fusion of several detectors for correct detection and recognition of attack types. The experiments held on both KDD’99 data and real network traffic data show promising results.