Multi-Scale Convolutional Neural Networks optimized by elite strategy dung beetle optimization algorithm for encrypted traffic classification

Quan Peng, Xingbing Fu, Fei Lin, Xiatian Zhu, Jianting Ning, Fagen Li

Abstract: Highlights•Design an improved MSCNN model for encrypted traffic classification.•Propose the ESDBO algorithm which finds better hyperparameter configurations.•Obtain the accuracy of 83.44% for the MSCNN model and 78.68% for the CNN model.•Our model accuracy has an average lead of 5.56% on MSCNN and 4.19% on CNN.