Proactive Cyber Defense: User-Centric Risk Assessment in DNS Traffic Through Graph-Based Learning
Bibliographic record
Abstract
In today’s rapidly evolving digital landscape, cybersecurity remains a paramount concern. Effectively fortifying network and internet defenses requires a deep understanding of infrastructure vulnerabilities and emerging threats. This paper introduces a sophisticated machine learning approach tailored to assess and enhance security through proactive cyber defense, specifically by profiling users within the context of Domain Name System (DNS) traffic analysis. Our method involves extracting user-domain browsing data and constructing a domain similarity graph, a representation that captures threat similarities among requested domains. We further present a graph-based anomalous risk assessment mechanism, augmented by a machine learning feedback loop. This innovative mechanism meticulously profiles users based on their online activities, assigns risk scores, and adeptly detects potential security threats. To validate the effectiveness of our approach, we leverage real-world DNS query data obtained from the Canadian Internet Registration Authority (CIRA), encompassing user-generated requests, query logs, responses, and a repository of high-threat domains. Our empirical evaluations demonstrate remarkable accuracy, achieving confidence levels of 96.8% and 98.2% in analyzing anomalous behavior among domains and users, respectively, using the provided dataset.
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 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.003 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 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".