HiNFRA: Hierarchical Neuro-Fuzzy Learning for Online Risk Assessment

Our previous research illustrated the design of fuzzy logic based online risk assessment for Distributed Intrusion Prediction and Prevention Systems (DIPPS) [3]. Based on the DIPPS sensors, instead of merely preventing the attackers or blocking traffic, we propose a fuzzy logic based online risk assessment scheme. This paper propose a Hierarchical Neuro-Fuzzy online Risk Assessment (HiNFRA) model to aid the decision making process of a DIPPS. The fine tuning of fuzzy logic based risk assessment model is achieved using a neural network learning technique. Preliminary results indicate that the neural learning technique could improve the fuzzy controller performance and make the risk assessment model more robust.