MétaCan
Menu
Back to cohort
Record W4414015728 · doi:10.11159/cist25.128

Identifying Malicious Network Traffic Detection using Graph Transformers & Masked Autoencoders

2025· article· en· W4414015728 on OpenAlexvenueno aff
Michael Choi

Bibliographic record

VenueProceedings of the World Congress on Electrical Engineering and Computer Systems and Science · 2025
Typearticle
Languageen
FieldComputer Science
TopicNetwork Security and Intrusion Detection
Canadian institutionsnot available
Fundersnot available
KeywordsComputer scienceTransformerComputer networkArtificial intelligencePattern recognition (psychology)Computer securityEngineeringElectrical engineeringVoltage

Abstract

fetched live from OpenAlex

The rise in cyber-attacks highlights the critical need for advanced network intrusion detection systems.Traditional machine learning methods often fail to capture the complex patterns inherent in cybersecurity data.Graph Neural Networks (GNNs) [4], capable of efficiently modeling data as nodes and edges, have shown promise in addressing these challenges.This research proposes a novel approach combining Graph Masked Autoencoder (Graph MAE) [2] for self-supervised pretraining and a global attention-based Graph Transformer (Graph GPS) [3] for fine-tuning.Utilizing the UNSW-NB15 dataset [1], we sampled 25% of the dataset (approximately 653,012 network flow records) due to computational restraints.Performance metrics such as Accuracy, Precision, Recall, F1-score, and Area Under the ROC Curve (AUC) were employed.Results indicate significant performance improvements (Accuracy: 0.95, Precision: 0.58, Recall: 0.94, F1-score: 0.72, AUC: 0.98) compared to a baseline two-layer Graph Convolution Network (GCN) [4] model.The study underscores the efficacy of combining self-supervised learning methods and global attention mechanisms in enhancing malicious 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 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)0.001
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.400
Threshold uncertainty score0.545

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0010.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.003
Science and technology studies0.0010.000
Scholarly communication0.0000.000
Open science0.0010.000
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 head0.009
GPT teacher head0.219
Teacher spread0.209 · 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

Classification

machine, unvalidated

Machine predicted; a candidate call from one teacher head, not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designSimulation or modeling
Domainnot available
GenreEmpirical

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".

Quick stats

Citations0
Published2025
Admission routes1
Has abstractyes

Explore more

Same venueProceedings of the World Congress on Electrical Engineering and Computer Systems and ScienceSame topicNetwork Security and Intrusion DetectionFrench-language works237,207