DoH detection method based on self-attention BiLSTM

The invention discloses a DoH detection method based on self-attention BiLSTM, and relates to the technical field of network information security, and the method comprises the steps: obtaining first traffic data; carrying out undersampling and oversampling processing on the first traffic data to obtain second traffic data; and inputting the second traffic data into a self-attention BiLSTM deep learning model, extracting global features of the DoH traffic data and retaining time sequence features. According to the invention, under-sampling and over-sampling are adopted to process flow data, so that data balance is realized; in order to accurately carry out multi-classification of DoH traffic, the self-attention BiLSTM deep learning model is constructed, and the problems that DoH traffic detection is difficult and the accuracy rate is low are solved. 本发明公开了一种基于自注意力BiLSTM的DoH检测方法，涉及网络信息安全技术领域，包括：获取第一流量数据；对所述第一流量数据进行欠采样和过采样处理，得到第二流量数据；将所述第二流量数据输入自注意力BiLSTM深度学习模型，提取所述DoH流量数据的全局特征并保留时序特征。本发明采用欠采样和过采样对流量数据进行处理，实现了数据