A DoH Traffic Detection System Based on Interpretable Deep Learning Neural Networks
Bibliographic record
Abstract
Over the past few years, DNS has been a preferred target for hacker attacks, and although the DNS over HTTPS (DoH) protocol is considered ideal for Internet users, providing additional privacy protection and security, it also poses a significant problem: network administrators are unable to effectively monitor and intercept suspicious network traffic generated by malware and malicious tools. In this paper, in order to address this need in the field of cybersecurity, we develop a model to accurately detect and classify DNS over HTTPS attacks using a deep learning neural network approach based on the Canadian Center for Secure Networking's open source CIRA-CIC-DoHBrw-2020 dataset. The developed model achieves more than 97% accuracy in the detection and classification task of comprehensive traffic and very high accuracy (99. 1 %), recall (98. 1 %), and F1 score (95.8%) in the scenario classification task of malignant DoH traffic and benign DoH traffic identification. At the end of this study, according to the concept of interpretable AI methods, the potential feature contributions of the model are computed and emphasized by the novel use of game theory-based SHAP method, which enables the developed algorithm to provide transparent and interpretable results, which will be useful and instructive for future work on malignant traffic detection.
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 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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".