Traffic data distribution problem and novel network attack pose great threat to the traditional machine learning based anomaly network traffic detection system. In this paper, we design a method based on deep transfer learning to try to solve these problems. To evaluate the performance of the proposed method, we use the basic classifiers KNN, SVM, RandomForest, Xgboost and the basic classifiers above based on the TCA mapping method as benchmark on the NSL-KDD dataset. The experiment result shows that it can solve the inconsistent distribution of different network traffic data and possible novel attacks in network traffic detection tasks to some extent.
展开▼