Research Article
Using Generative Adversarial Networks for Network Intrusion Detection
@INPROCEEDINGS{10.1007/978-3-031-04245-4_6, author={XuDong Li and Di Lin and Yu Tang and Weiwei Wu and Zijian Li and Bo Chen}, title={Using Generative Adversarial Networks for Network Intrusion Detection}, proceedings={6GN for Future Wireless Networks. 4th EAI International Conference, 6GN 2021, Huizhou, China, October 30--31, 2021, Proceedings}, proceedings_a={6GN}, year={2022}, month={5}, keywords={Generative adversarial network Network intrusion detection Network security}, doi={10.1007/978-3-031-04245-4_6} }- XuDong Li
Di Lin
Yu Tang
Weiwei Wu
Zijian Li
Bo Chen
Year: 2022
Using Generative Adversarial Networks for Network Intrusion Detection
6GN
Springer
DOI: 10.1007/978-3-031-04245-4_6
Abstract
The network intrusion detection system is an essential guarantee for network security. Most research on network intrusion detection systems focuses on using supervised learning algorithms, which require a large amount of labeled data for training. However, the work of labeling data is complex and cannot exhaustively include all types of network intrusion. Therefore, in this study, we develop a model that only requires normal data in the training phase, and it can distinguish between normal data and abnormal data in the test phase. This model is implemented by using a generative confrontation network. Experimental results show that, on the CIC-IDS-2017 dataset, our model has an accuracy of 97%, which is dramatically higher than the basic autoencoder, which is one of the most widely used algorithms in the network intrusion detection.
