MétaCan
Menu
Back to cohort

Application of Machine Learning to Determine Cybersecurity Compliance: Analyze the Risk of Privacy Violations in Distributed Systems

2025· article· W7128727797 on OpenAlex

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

affAt least one author lists a Canadian institution in the pinned OpenAlex snapshot.

Bibliographic record

Venuenot available
Typearticle
Language
FieldComputer Science
TopicSoftware System Performance and Reliability
Canadian institutionsArtificial Intelligence in Medicine (Canada)
Fundersnot available
KeywordsScalabilityAnomaly detectionProcess (computing)Information privacyServerDistributed database

Abstract

fetched live from OpenAlex

Distributed computing environments grow exponentially – encompassing cloud, edge, and IoT ecosystems. This trend leads to near-unbeatable complexity in cybersecurity compliance management. Organizations deploying distributed systems around cloud, edge, and IoT for scalability and real-time operons find it more and more complicated to guarantee that systems comply with privacy regulations, including GDPR, HIPAA, CCPA, and CL In distributed architectures, manual auditing, and other rule-based options fell short on detecting emerging threats or concealed risks, as well as delicate deviations from a policy across a vast number of nodes. ML-based techniques could automate and significantly improve the process of ensuring cybersecurity compliance and analyzing privacy risks in completive distributed mechanisms. The approach described in this paper relies upon supervised and unsupervised ML models considering patterns of system operations, user accesses, and configurations across distributed environments. Thorough registered training on compliance, the models classify potential privacy violatio9ns, measure the alignment with compliance policies, and generate futuristic predictions on the matter of non-compliance. Functionalities like dynamic profiling, real-time correlations, and anomaly detection assess a system’s compliance posture, prioritizing most relevant and severe risks. The explainability modules have been included, allowing interpreting system decisions and facilitating human-in-the-loop decisions. The developed approach has been assessed in experiments with hybrid datasets containing synthetic and real-life logs from the multicloud scenarios. According to the results, the approach demonstrates high accuracy of classification, robust anomaly detection, and the possibility of early risk detection, largely exceeding the performance of traditional rule-based systems. The system also required minor adjustments to comply with novel rules, making it compatible for organizations facing dynamic legislative environments. All-in-all, this paper confirms the hypothesis that compliant analysis powered by ML can bridge the gap between static policy documents and dynamic distributed architectures. Making data produce actionable insights on compliance not only improves cybersecurity postures but also allows stakeholders to report to regulators and control risks more efficiently. The research provided a basis for future work in autonomous compliance systems development and privacy-aware AI governance frameworks.

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.

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.002
metaresearch head score (Gemma)0.001
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Observational · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.650
Threshold uncertainty score0.998

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0020.001
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0010.000
Bibliometrics0.0000.004
Science and technology studies0.0000.000
Scholarly communication0.0000.000
Open science0.0010.001
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

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.

Opus teacher head0.015
GPT teacher head0.279
Teacher spread0.264 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it