A Network Intrusion Detection Method Based on Hybrid Rice Optimization Algorithm Improved Fuzzy C-Means

摘要

Fuzzy C-means (FCM) is a classical clustering method for data analysis in the machine learning field which has been successfully applied in the network intrusion detection. But it is sensitive to the initial clustering center, isolated point and noise, as well it is easy to trap into the local optimal solution. There are some methods for alleviating this problem by using optimization algorithms such as genetic algorithm, particle swam optimization algorithm, grey wolf optimization algorithm etc. However, in general, there are multiple local optimal values in the objective function of FCM, which is not fully conquered with evolutionary algorithms. A newly proposed hybrid rice optimization algorithm is employed to improve the basic FCM(HROFCM) and applied for network intrusion detection. Finally, KDD'99 data is used to test the effectiveness of the method and experimental results show that the performance of HROFCM prevails in the comparison of GAFCM(fuzzy c-means based on genetic algorithm) and GWOFCM (fuzzy c-means based on grey wolf optimization algorithm).