| Popis: |
Accurate identification of network attacks has become an important topic in the field of network security, and the use of machine learning methods to identify network attacks has been one of the research hotspots in the field of network security protection in recent years. But they have some limitations, such as model parameters are difficult to optimize, and the prediction accuracy rate is low. Also, the characteristics of time series uncertainty and nonlinearity in network traffic messages increase the difficulty of forecasting. This article proposes a prediction model using the CNN+LSTM network and uses feature engineering methods to preprocess traffic sample data, and then select effective features through information gain and determining feature weights. Use the NSL-KDD data set to conduct empirical research on the prediction algorithm based on feature engineering and long-short-term memory artificial neural network proposed in this paper. The results show that the prediction model based on CNN+LSTM is compared with only LSTM algorithm or only classification algorithm. It can improve prediction accuracy, reduce model training time, and is easier to apply in the actual work of network attack identification and security prevention. |
