Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Illicit Bitcoin transaction detection via feature-gated temporal graph learning.

Na Han1, Ruixian Zhang1, Xiaoyun Liu1

  • 1Shandong Vocational and Technical University of International Studies, Rizhao, 276826, China.

Scientific Reports
|May 17, 2026
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Liposomal bupivacaine in peri-articular injection reduces opioid consumption and improves early prognosis of total knee arthroplasty: a randomized controlled trial.

BMC musculoskeletal disorders·2026
Same author

Aseptic meningitis as an uncommon presentation of polyarteritis nodosa: a case report with focused literature synthesis.

BMC neurology·2026
Same author

Quantifying Short-Term Functional Changes After Lecanemab Treatment in Early Alzheimer's Disease: An Exploratory 3-Month Follow-Up Case Report Using Eye Movement and Gait Analysis.

Current Alzheimer research·2026
Same author

Limited ability of coronal alignment parameters to assess rotational alignment in patients with varus knee osteoarthritis: a cross-sectional study.

BMC musculoskeletal disorders·2026
Same author

Analysis of Appressorium Formation in <i>Metarhizium anisopliae</i> and Its Impact on the Defense Metabolism of <i>Opisina arenosella</i> Larvae Based on LC-MS.

Insects·2026
Same author

Enhanced PEMFC Performance and Durability Enabled by a Highly Oxygen-Permeable Ionomer Binder.

ACS omega·2026
Same journal

MT-MRI for detection of renal interstitial fibrosis in renovascular disease.

Scientific reports·2026
Same journal

Detection of underground objects from GPR data using a lightweight YOLO-based approach.

Scientific reports·2026
Same journal

Early systemic inflammatory-metabolic trajectory phenotypes are associated with survival outcomes in metastatic renal cell carcinoma treated with nivolumab.

Scientific reports·2026
Same journal

Water balance components in a dry-seeded rice-wheat system: Untangling the effects of tillage and mulching practices.

Scientific reports·2026
Same journal

Topological approaches to quantum tensor train compression via ZX-calculus and SVD.

Scientific reports·2026
Same journal

determinants of flood impacts and adaptive capacity among market vendors in Walukuba-Masese, Jinja city, Uganda.

Scientific reports·2026
See all related articles

We developed FG-EGCN, a novel model for detecting illicit Bitcoin transactions. It effectively balances temporal graph data and transaction features, improving detection accuracy, especially for rare illicit activities in imbalanced datasets.

Area of Science:

  • Blockchain Technology and Cybersecurity
  • Machine Learning and Graph Neural Networks
  • Financial Forensics and Anti-Money Laundering

Background:

  • Illicit transaction detection in blockchains is crucial for anti-money laundering efforts.
  • Existing graph-based methods struggle with temporal dynamics, class imbalance, and attribute preservation.
  • Performance degrades with unstable graph structures and weak local context for illicit nodes.

Purpose of the Study:

  • To propose FG-EGCN, a feature-gated temporal graph model for robust illicit Bitcoin transaction detection.
  • To enhance the balance between temporal-relational evidence and transaction-level attribute information.
  • To improve sensitivity to rare illicit samples in highly imbalanced datasets.

Main Methods:

  • Developed a feature-gated temporal graph convolutional network (FG-EGCN) model.
Keywords:
Bitcoin transaction graphBlockchain anti-money launderingClass imbalanceFeature-gated fusionGraph neural networksIllicit transaction detectionTemporal graph learning

Related Experiment Videos

  • Integrated temporal graph encoding with a residual feature branch and adaptive gating.
  • Employed focal-loss optimization for imbalanced learning and chronological train-test split for evaluation.

    • Main Results:

    • FG-EGCN demonstrated strong performance and robustness against temporal distribution shifts on the Elliptic dataset.
    • The model achieved improved temporal stability and confirmed the contributions of its core components via ablation studies.
    • Visualizations revealed a more structured, interpretable embedding space with better separation of illicit nodes.

    Conclusions:

    • Combining temporal graph learning with feature preservation and adaptive fusion offers an effective solution for illicit Bitcoin transaction detection.
    • FG-EGCN provides a more robust and interpretable approach compared to existing graph neural network models.
    • The model's design addresses key challenges including temporal evolution, feature importance, and data imbalance.