Imbalanced Encrypted Traffic Classification Scheme Using Random Forest
Feng Zhang, Tao Shang, Jianwei Liu · 2020
Encrypted traffic classification techniques can identify different types of traffic for network security. The existing schemes seldom consider the imbalanced distribution of the traffic. In this paper, we propose an encrypted traffic classification scheme using random forest for imbalanced learning. Firstly, the weighted information gain is used to select the features which are beneficial to the minority class to filter redundant features. Then the hybrid sampling method is used to balance the number of the majority class and the minority class. Finally, the random forest is constructed for encrypted traffic classification. Experimental results show that the feature selection method can filter the redundant features of the traffic and the hybrid sampling method can effectively improve the classification ability. Compared with K-nearest neighbor and C4.5 decision tree algorithm, the proposed scheme can classify imbalanced encrypted traffic more effectively.