Machine-learning and statistical methods for DDoS attack detection and defense system in software defined networks
Bibliographic record
Abstract
Distributed Denial of Service (DDoS) attack is a serious threat on today’s Internet. As the traffic across the Internet increases day by day, it is a challenge to distinguish between legitimate and malicious traffic. This thesis proposes two different approaches to build an efficient DDoS attack detection system in the Software Defined Networking environment. SDN is the latest networking approach which implements centralized controller, which is programmable. The central control and the programming capability of the controller are used in this thesis to implement the detection and mitigation mechanisms. In this thesis, two designed approaches, statistical approach and machine-learning approach, are proposed for the DDoS detection. The statistical approach implements entropy computation and flow statistics analysis. It uses the mean and standard deviation of destination entropy, new flow arrival rate, packets per flow and flow duration to compute various thresholds. These thresholds are then used to distinguish normal and attack traffic. The machine learning approach uses Random Forest classifier to detect the DDoS attack. We fine-tune the Random Forest algorithm to make it more accurate in DDoS detection. In particular, we introduce the weighted voting instead of the standard majority voting to improve the accuracy. Our result shows that the proposed machine-learning approach outperforms the statistical approach. Furthermore, it also outperforms other machine-learning approach found in the literature.
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 | 0.002 | 0.007 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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".