An Intrusion Detection Model Based on Improved Whale Optimization Algorithm and XGBoost
Xinlu Zong, Ruicheng Li, Zhiwei Ye · 2021 11th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications (IDAACS) · 2021
Hyperparameters optimization is an effective way to improve the detection performance of intrusion detection models based on machine learning. However, hyperparameters optimization is a typical NP-hard problem. Therefore, it is difficult to find the optimal parameters within an acceptable time by using traditional parameters optimization methods. To solve this problem, this paper proposed an intrusion detection model named IWOA-XGB which combines improved Whale Optimization Algorithm (WOA) and eXtreme Gradient Boosting (XGBoost). Firstly, the spiral update maneuver of the original WOA is modified by setting the parameter$l$to a linearly decreasing random value. Then, the improved WOA is used to optimize several valued parameters of tree booster in the XGBoost. Finally, the optimized XGBoost is applied for intrusion detection so that the performance of the intrusion detection model can be improved. To evaluate the efficacy of IWOA-XGB, the NSL-KDD intrusion detection data set is used for simulation tests. The simulation results show that compared to the original XGBoost and other evolutionary algorithms, IWOA-XGB could effectively improve the macro F1-score of the detection model, demonstrating good intrusion detection performance.