基于集成RBF神经网络的入侵检测研究

Intrusion Detection Research Based on the Integration of RBF Neural Network

本文提出了一种基于集成RBF神经网络的入侵检测研究,通过对初始化聚类子算法的改进,从而提高了RBF神经网络的训练速度,采用集成理论对RBF神经网络的集成以提高检测率。实验结果表明,集成神经网络比RBF神经网络的检测率提高了1%,且降低了误报率和漏报率。

This paper proposes an intrusion detection research based on the integration RBF neural network. By improving the initialization sub-clustering algorithm, the training speed of the RBF can be improved. By integrating some differential RBF neural network based on the integrated theory, the detection rate is improved. The experimental results show that compared with the detection rate of the RBF, that of the integrated neural network is increased by 1%, and the latter one reduces the false positive rate and the false negative rate.