Intrusion Detection System based on GRU
Bayi Xu, Lei Sun, Ruiyang Ding, Chengwei Liu · 2023
In recent years, there has been a rise in network attack incidents. Efficiently detecting network attack behaviors is crucial for ensuring cybersecurity in resource-limited environments. This paper proposes a lightweight intrusion detection system based on GRU (Gated Recurrent Unit). The system uses information gain and Extra Tree Classifier methods to rank and select important features. Temporal features of network traffic are extracted using gated recurrent neural networks, and classification is performed using the softmax function. To validate its effectiveness, the NSL-KDD dataset is used for evaluation, and experimental results show that the proposed model outperforms other state-of-the-art models.