Generation of Similar Traffic Using GAN for Resolving Data Imbalance
Recently, as the practical application of deep learning has become possible, research on the problems pertaining to intrusion detection has increased.
However, it is difficult to detect a small number of attack traffic when the real network is connected to produce an imbalance between the attack traffic class data and the normal traffic data necessary for learning. In this study, we propose a method to improve the accuracy of attack traffic data detection by creating similar attack traffic, using a Generative Adversarial Network (GAN) algorithm of deep learning. The proposed method generates similar attack traffic for NSL–KDD, ISCX 2012, and USTC_TFC 2016 datasets, which are well-known intrusion detection learning data sets. Experiments have shown that the data imbalance in each data set can improve classification accuracy by 10–12%, owing to the degradation problem.
KeywordsDeep learning Intrusion detection Security Generative Adversarial Network
- 1.Tavallaee, M., Bagheri, E., Lu, W., Ghorbani, A.: A detailed analysis of the KDD CUP 99 data set. Submitted to Second IEEE Symposium on Computational Intelligence for Security and Defense Applications (CISDA) (2009)Google Scholar
- 3.Wang, W., et al.: Malware traffic classification using convolutional neural network for representation learning. In: 2017 International Conference on Information Networking (ICOIN). IEEE (2017)Google Scholar
- 6.Tang, T.A., et al.: Deep learning approach for network intrusion detection in software defined networking. In: 2016 International Conference on Wireless Networks and Mobile Communications (WINCOM). IEEE (2016)Google Scholar
- 7.Mylavarapu, G., Thomas, J., Ashwin Kumar, T.K.: Real-time hybrid intrusion detection system using apache storm. In: High Performance Computing and Communications (HPCC), 2015 IEEE 7th International Symposium on Cyberspace Safety and Security (CSS), 2015 IEEE 12th International Conference on Embedded Software and Systems (ICESS), 2015 IEEE 17th International Conference on. IEEE (2015)Google Scholar
- 10.Tavallaee, M., et al.: A detailed analysis of the KDD CUP 99 data set. In: IEEE Symposium on Computational Intelligence for Security and Defense Applications, CISDA 2009. IEEE (2009)Google Scholar
- 11.Hariharan, B., Girshick, R.: Low-shot visual recognition by shrinking and hallucinating features. In: Proceedings of IEEE International Conference on Computer Vision (ICCV), Venice, Italy (2017)Google Scholar
- 12.DCGAN: Radford, A., Metz, L., Chintala, S.: Unsupervised representation learning with deep convolutional generative adversarial networks. arXiv preprint arXiv:1511.06434 (2015)