An Efficient Intrusion Detection Model Based on Convolutional Neural Network and Transformer
ZheYing Zhang, Lijuan Wang
Abstract
With the increase of network traffic and the highly uneven distribution of data, it is very difficult to make the boundary between normal behavior and abnormal behavior. The openness and variability of network and its services make the threat space increasing. Therefore, effective intrusion detection models are needed to adapt to the dynamic environment and requirements. We proposed a intrusion detection model which combining the convolutional neural network and Transformer together. The presented model can not only capture the global correlation between data packets, but also the local correlation of an intrusion. The experimental results show that our model can improve the detection accuracy and decrease the training time.