An ensemble learning-based two-level network intrusion detection method
Jianxin Zheng, Xuming Ni, Lifeng Li, Kan Yu, Jun Zhang · 2022 International Conference on Computer Engineering and Artificial Intelligence (ICCEAI) · 2022
Nowadays, network security has become a serious issue as network attacks have increased significantly. Intrusion detection system (IDS) is an effective means to detect network attack in real time. However, current method has the low accuracy and a long detection time for multiple attacks detection. To solve these problems, we propose a two-level network intrusion detection method, which is based on ensemble learning. The first-level uses a binary-classification model to quickly determine whether the network access behavior is an attack. The second-level adopts a multi-classification model, which can classify anomalous access behaviors into specific attack types. Some experiments were taken to validate the effectiveness of our method. The experimental results show that our method has 99.91% on accuracy, low false alarm rate and a fast detection speed, which is superior to state-of-the-art methods.