中圖分類號：TP393文獻標識碼：ADOI:10.19358/j.issn.2097-1788.2025.07.003
引用格式：胡宏偉，孫皓月. 基于超參數(shù)優(yōu)化和LightGBM算法的DDoS攻擊檢測與分類［J］.網(wǎng)絡(luò)安全與數(shù)據(jù)治理，2025，44（7）：15-19，26.

DDoS attack detection and classification based on hyperparameter optimization and LightGBM algorithm

Abstract： Aiming at the characteristics of large sample capacity and multiple data features of distributed denial of service attack (DDoS) data traffic as well as the problem of low detection and classification accuracy, this paper proposes a DDoS attack detection and classification method based on LightGBM (Light Gradient Boosting Machine) algorithm.Based on the preprocessing and feature screening of the CICDDoS2019 dataset, the LightGBM detection model and multi-classification model are constructed. Meanwhile, random grid search and Bayesian hyperparameter optimisation techniques are used to achieve hyperparameters auto-tuning during model pre-training.The experimental results show that the model in this paper can achieve an accuracy rate of 98.34% in the detection and classification tasks. This research aims to provide an efficient and simple detection and classification idea for DDoS attacks.

Key words : DDoS attacks; hyperparameter optimization; LightBGM; detection and classification