Intrusion Detection System Using Kernel FCM Clustering and Bayesian Neural Network

Data safekeeping and security has been a key concern in the rapidly growing computer systems and networks. One of the recent methods for identifying any abnormal activities staging in a computer system is carried out by Intrusion Detection Systems (IDS) and it forms a significant portion of system defence against attacks. Various methods based on Intrusion Detection Systems have been proposed in recent years. In this paper, Intrusion Detection System (IDS) based on Fuzzy BisectorKernel Fuzzy C-means clustering technique and Bayesian Neural Network is proposed. The system contains two steps namely clustering step and classification step. In clustering step, the input dataset is grouped into clusters with the use of Fuzzy BisectorKernel Fuzzy C-means clustering (FB-KFCM). In the classification step, the centroids from the clusters are taken for training in the Bayesian Neural Network. Subsequently, test data is given to the trained network, which gives the outputs if the data is intruded or not. The proposed technique is implemented by JAVA PROGRAMMING using KDD CUP 99 dataset. The evaluation metric utilized is accuracy and comparative analysis is made to other techniques. The average accuracy value obtained is 93.91which was better than other compared techniques. The high accuracy value shows the efficiency of the proposed technique.