From Real Malicious Domains to Possible False Positives in DGA Domain Detection
Bibliographic record
Abstract
Various families of malware use domain generation algorithms (DGAs) to generate a large number of pseudo-random domain names to connect to malicious command and control servers (C&Cs). These domain names are used to evade domain based security detection and mitigation controls such as firewall controls. Existing prevalent techniques to detect DGA domains such as reverse engineering malware samples and statistical analysis techniques are time consuming, can be easily circumvented by attackers, and need contextual information which is not easily or feasibly obtained. Due to this, the use of machine learning and deep learning techniques to detect DGA domains has picked up significant interest in the cyber security and analytics communities. The ultimate goal is to detect DGA domains on a per domain basis using the domain name only, with no additional information. As with all techniques, there is the possibility of false positives: valid domains being detected as DGA domains. This paper explores the possible use cases that can result in false positives for DGA domain detection using machine learning and deep learning techniques, and how such use cases, if not uniquely addressed within an automated system or model or technique, can also be used as attack vectors by attackers using DGA domains.
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 | 0.011 | 0.085 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.004 | 0.002 |
| Science and technology studies | 0.001 | 0.003 |
| Scholarly communication | 0.004 | 0.004 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.003 | 0.003 |
| 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".