FTG-Net-E: A hierarchical ensemble graph neural network for DDoS attack detection
Bibliographic record
Abstract
Distributed Denial-of-Service (DDoS) attacks are a major threat to computer networks. These attacks can be carried out by flooding a network with malicious traffic, overwhelming its resources, and/or making it unavailable to legitimate users. Existing machine learning methods for DDoS attack detection typically use statistical features of network traffic, such as packet sizes and inter-arrival times. However, these methods often fail to capture the complex relationships between different traffic flows. This paper proposes a new DDoS attack detection approach that uses Graph Neural Networks (GNN) ensemble learning. GNN ensemble learning is a type of machine learning that combines multiple GNN models to improve the detection accuracy. We evaluated our approach on the Canadian Institute for Cybersecurity Intrusion Detection Evaluation Dataset (CICIDS2018) and CICIDS2017 datasets, a benchmark dataset for DDoS attack detection. Our work provides two main contributions. First, we extend our DDoS attack detection approach using GNN ensemble learning. Second, we explore the evaluation and fine-tuning of hyperparameter metrics through ensemble learning, significantly enhancing accuracy compared to a single GNN model and achieving an average 3.2% higher F1-score. Additionally, our approach effectively reduces overfitting by incorporating regularization techniques, such as dropout and early stopping. Specifically, we use a hierarchical ensemble of GNN, where each GNN learns the relationships between traffic flows at a different granularity level. We then use bagging and boosting to combine the predictions of the individual GNN, further improving detection accuracy. Results show that our system can achieve 99.67% accuracy, with a F1-score of 99.29%, which is better than state-of-the-art methods, even using single traffic architecture.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".