The Research on Fuzzy Clustering Method Based on Differential Evolution Algorithm in Intrusion Detection
Qi Yang, Li Na Wang · Applied Mechanics and Materials · 2014
Fuzzy C-means clustering algorithm (FCM) is widely applied to the intrusion detection. To acquire a better division for intrusion data, a new method (DEFCM) presented in the paper which combines FCM and differential evolution algorithm (DE) is found application. As a start, several randomly initiated partitions are optimized by FCM, and then the result is provided to differential evolution algorithm. After that, the combined result is sent to FCM again to adjust the partition and obtain the final answer. The method can improve detection performance effectively. The KDDCUP1999 data set is used in the simulation experiment, and the result proves that the DEFCM algorithm has a comparatively high detection rate in intrusion detection.