A Fuzzy Neural Network and Multiple Kernel Fuzzy C-means Algorithm for Secured Intrusion Detection System

An Intrusion Detection System (IDS) is a security layer used to detect constant intrusive behavior in information systems. Many intrusion detection systems have been proposed based on the various data mining approaches such as decision tree, clustering, etc. Although the intrusion detection system is efficient way to find the attacks in the system, existing ones have some disadvantages which affects the performance of the system. It is observed that Neural Networks improves the overall performance of the intrusion detection system when it is integrated with a clustering approach. This research work aims to improve the performance of the intrusion detection system through the application of Fuzzy Neural Network along with an efficient fuzzy clustering method. In this proposed approach, initially Multiple Kernel Fuzzy C-Means (MKFCM) technique is used to construct different training subsets. The performance of the fuzzy clustering approach is improved through MKFCM. Then, different FNN models are trained to formulate different base models according to the different training models. Then the final results are aggregated through fuzzy based approach. The performance of the proposed MKFCM-FNN approach is compared with other existing approaches.