Scholarlink

The rapid expansion of internet-based systems has led to a massive increase in cyberattacks, particularly Distributed Denial of Service (DDoS), which is capable of causing significant operational and financial disruptions. In this study, we propose a hybrid of Machine Learning models called the Stacking ensemble. We employed Logistic Regression (LR), LightGBM and XGBoost models as the base learners and Logistic Regression (LR) as the meta learner model on the CIC-DDoS2019 and UNSW-NB15 datasets to analyze and obtain the evaluation metrics (accuracy, recall, precision, and F1 score) and compare them against those obtained in multiple research papers. Our model obtained 93.91% recall, 94.11% precision, and 93.72% F1-score, demonstrating its ability to detect DDoS attacks with high accuracy and minimal false positives.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Lightweight and Enhanced Machine Learning Based Ensemble Framework for DDoS Attack Detection

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Nathan D’Souza,
  • Yohaan Guzdar,
  • Ruchi Sharma

摘要

The rapid expansion of internet-based systems has led to a massive increase in cyberattacks, particularly Distributed Denial of Service (DDoS), which is capable of causing significant operational and financial disruptions. In this study, we propose a hybrid of Machine Learning models called the Stacking ensemble. We employed Logistic Regression (LR), LightGBM and XGBoost models as the base learners and Logistic Regression (LR) as the meta learner model on the CIC-DDoS2019 and UNSW-NB15 datasets to analyze and obtain the evaluation metrics (accuracy, recall, precision, and F1 score) and compare them against those obtained in multiple research papers. Our model obtained 93.91% recall, 94.11% precision, and 93.72% F1-score, demonstrating its ability to detect DDoS attacks with high accuracy and minimal false positives.