Network Traffic Anomaly Detection Method Based on S-UBayFS Feature Selection
We study the method of network traffic anomaly detection,and propose an algorithm based on S-UBayFS-GRU to improve the limitations of traditional machine learning methods.The algorithm is divided into three steps.It uses SNHA algo-rithm to filter out causally related features from a large number of network traffic features and form a"correlation chain".It u-ses the"correlation chain"and network security domain knowledge to assign weights and side constraints to the features.It al-so uses the"correlation chain"and network security domain knowledge,and uses UBayFS algorithm for feature selection to re-duce the feature dimensions and improve the feature quality.GRU recurrent neural network is used to learn and predict the fil-tered features,and realize the network traffic anomaly detection.Experiment results show that the S-UBayFS-GRU algorithm proposed in this paper outperforms other methods in all evaluation indexes.
万方数据
维普
