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

Advanced behavioral malware detection: a comprehensive MLOps framework with federated learning and real-time drift

Mohammed El-Hajj1, Mohammad Al Jawad Zeineddine2

  • 1Faculty of Computer Studies (FCS), Arab Open University (AOU), Beirut, Lebanon.

Frontiers in Artificial Intelligence
|May 27, 2026
PubMed
Summary

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

Explainable AI-driven customer churn prediction: a multi-model ensemble approach with SHAP-based feature analysis.

Frontiers in artificial intelligence·2026
See all related articles

This study introduces a robust MLOps framework for behavioral malware detection, enhancing generalization and operational resilience. It features novel validation, efficient feature engineering, and privacy-preserving federated learning for improved cybersecurity.

Area of Science:

  • Cybersecurity
  • Machine Learning Operations (MLOps)

Background:

  • Behavioral malware detection faces challenges in generalization, collaboration, and operational resilience.
  • Existing methods often lack robust validation and efficient real-time adaptation.

Purpose of the Study:

  • To present a comprehensive MLOps framework for behavioral malware detection.
  • To address generalization, collaboration, and operational resilience challenges.
  • To bridge the gap between academic research and operational cybersecurity.

Main Methods:

  • Formalized Leave-One-Experiment-Out (LOEO) validation protocol for conservative generalization assessment.
  • Domain-optimized feature engineering pipeline for hierarchical behavioral signatures.
  • Hybrid federated learning architecture for privacy-preserving collaboration with differential privacy guarantees.
Keywords:
LOEO cross-validationLightGBM optimizationbehavioral malware detectionconcept drift adaptationfederated learningreal-time threat detection

Related Experiment Videos

  • Real-time concept drift detection engine triggering automated retraining.

    • Main Results:

    • LOEO validation revealed a 12.3% accuracy drop compared to conventional evaluation.
    • Feature engineering maintained 99.2% accuracy with 50% reduced inference latency.
    • Federated learning achieved 75.1% accuracy with (ϵ, δ)-differential privacy.
    • Drift detection and retraining achieved total recovery time < 5 min.
    • Evaluated on 2.74 million samples across 104 malware experiments with 10,000+ events/s throughput.

    Conclusions:

    • The proposed MLOps framework significantly enhances behavioral malware detection capabilities.
    • The framework demonstrates improved generalization, privacy-preserving collaboration, and operational resilience.
    • This production-oriented implementation provides a viable solution for operational cybersecurity requirements.