CS-WV Integrated Learning Model for Network Traffic Anomaly Detection
Yukun Li · 2023
As Internet technology develops rapidly, the security, validity and authenticity of network data have received more and more attention and importance. In order to maintain network security and accurately detect abnormal network traffic data, the study attempts to use integrated learning models for abnormality detection and classification, while improving on the basis of the weighted voting algorithm, and designs a classifier selection - weighted voting (CS-WV) integrated learning model based on The CS-WV integrated learning model is used to detect network traffic anomalies. The performance test results show that the curve area AVC of the CS-WV integrated learning model is 0.98, and the Precision, Accuracy, F1-Score and Recall of the CS-WV integrated learning model reach 98.70%, 98.93%, 96.07% and 93.59% respectively. integrated learning model improved Accuracy by 3.86%, Accuracy by 6.94%, F1-Score by 9.96% and Recall by 13.55% over the average accuracy rate. The above results show that the study of network traffic anomaly detection model (NTAD) using CS-WV integrated learning model has some reference value.