期刊名称：International Journal of Security and Its Applications
印刷版ISSN：1738-9976
出版年度：2016
卷号：10
期号：4
页码：423-430
DOI：10.14257/ijsia.2016.10.4.37
出版社：SERSC
摘要：To detect various network attacks in real time, this paper developed a network intrusion detection system based on artificial neural network. This paper first introduced the recent development of neural network, BP algorithm and structure of a simple perceptron. Then, this paper developed an improved BP neural network algorithm to detect anomaly network traffic with adjusted correlation rules. Finally, the network intrusion system in this paper was tested in a real network situation; the improved BP algorithm neural network with adjusted correlation rules shows a reduction in total error and increment in alarm rate compared to the traditional basic BP algorithm model.
