Network Intrusion Detection Model based on Fuzzy Support Vector Machine

Network intrusion detection is of great importance in the research field of information security in computer networks. In this paper, we concentrate on how to automatically detect the network intrusion behavior utilizing fuzzy support vector machine. After analyzing the related works of the proposed paper, we introduce the main characterics of fuzzy support vector machine, and demonstrate its formal description in detail. Next, the proposed intrusion detection system is organized as five modules, which are Data source, AAA protocol, FSVM module located in local computer, Guest computer and Terminals. Particularly, the intrusion detection module is constructed by four sections, which are data gathering section, data pre-processing section, intrusion detecting section and decision response section. Then, the intrusion detection algorithm based on fuzzy support vector machine is implemented by training process and testing process. Utilizing this algorithm, a sample in testing data can be judged whether it is belonged to network intrusion behavior. Finally, experimental results verify the effectiveness of our method comparing with other methods under different metric.