A neutral network classifier based design support system (NNCDSS) for network intrusion detection and response
TP Tran, Jianguo Lü, Donghua Wang, Shu‐Fang Chen, Matthew E. Tolentino · UTS ePRESS (University of Technology Sydney) · 2006
This paper introduces an innovative design of a classifier based decision support component for an intrusion detection system.In particular, this model uses an emerging semi-parametric learning algorithm called Modified Probabilistic Neural Network to capture both attacks' signatures as well as normal system usage behaviours.The statistics from this detection system is then assessed by an Expert System to generate a risk level and recommended actions in response to the detected anomalies.Extensive experimental outcome shows that the proposed model outperforms existing methods in terms of detection accuracy, model robustness and computational cost.