According to network traffic classification, because the previous network traffic classification method is single machine learning classification, it is difficult to improve its Overall Accuracy and this problem exists for a long time. Therefore, this paper proposes a new network traffic classification method, this method is based on machine learning classification, combines the advantages of different classification methods, uses integrated learning, and obtains the classification of network traffic through weighted average combination. The experiment shows that this method raises Overall Accuracy and is better than single machine learning classification.