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

A hybrid machine learning and explainable AI framework for optimizing risk-based authentication.

K Sasikumar1, Sivakumar Nagarajan1

  • 1School of Computer Science and Engineering, Vellore Institute of Technology, Vellore, India.

Plos One
|May 26, 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

Functional Outcome of Distal Third Tibial Fractures Treated by Posterior versus Medial Plating - A Prospective Comparative Study.

Journal of orthopaedic case reports·2025
Same author

The role of ABI2 in modulating nuclear proteins: Therapeutic implications for NUP54 and NUP153 in TNBC.

Advances in protein chemistry and structural biology·2025
Same author

In silico network pharmacology analysis and molecular docking validation of Swasa Kudori tablet for screening druggable phytoconstituents of asthma.

Advances in protein chemistry and structural biology·2024
Same author

Genomic insights and advanced machine learning: characterizing autism spectrum disorder biomarkers and genetic interactions.

Metabolic brain disease·2023
Same author

3D CoMoO<sub>4</sub> nanoflake arrays decorated disposable pencil graphite electrode for selective and sensitive enzyme-less electrochemical glucose sensors.

Mikrochimica acta·2022
Same author

Publisher Correction: Graphene overcoats for ultra-high storage density magnetic media.

Nature communications·2021
Same journal

Analysis of strength degradation of coal and rock masses and stability of mined areas under long term immersion environment.

PloS one·2026
Same journal

Biogenic Silver-Selenium nanocomposite with anticancer activity and potent efficacy against vancomycin-resistant Staphylococcus aureus.

PloS one·2026
Same journal

Preparation and physicochemical characterization of a biodegradable chitosan/carboxymethyl cellulose hydrogel synthesized in NaOH/urea medium.

PloS one·2026
Same journal

Action-guilt, survivor-guilt, and depression in combat-related PTSD.

PloS one·2026
Same journal

Explainable machine learning for predicting activities of daily living at discharge in stroke patients: A retrospective study using SHAP interpretability.

PloS one·2026
Same journal

Deep learning based two-way feature depiction model for brain tumor detection.

PloS one·2026
See all related articles

This study introduces a hybrid Risk-Based Authentication (RBA) framework using machine learning and fuzzy logic to enhance account takeover detection. The adaptive system improves security and user experience by dynamically assessing login risks.

Area of Science:

  • Cybersecurity
  • Artificial Intelligence
  • Machine Learning

Background:

  • The increasing growth of online platforms necessitates robust authentication methods to safeguard sensitive data and networks.
  • Risk-Based Authentication (RBA) offers an adaptive security approach, dynamically adjusting authentication based on user behavior and context to balance security and user experience.

Purpose of the Study:

  • To propose a novel hybrid Risk-Based Authentication (RBA) framework.
  • To enhance account takeover detection and dynamic risk assessment by integrating machine learning ensemble techniques, fuzzy logic, clustering, and optimization.

Main Methods:

  • An ensemble classifier (Gradient Boosting, SVM, XGBoost) was developed to predict account compromise probability using login behavior, device, and network data.
  • K-Means clustering generated initial risk thresholds, refined by a fuzzy logic system for probability-to-risk level mapping.

Related Experiment Videos

  • The L-BFGS-B optimization algorithm fine-tuned fuzzy membership boundaries for improved threshold consistency.

    • Main Results:

    • The framework achieved high performance metrics: 97.77% accuracy, 99.41% precision, 98.04% recall, 98.72% F1-score, and an Equal Error Rate (EER) of 0.0303.
    • On large datasets (2M-30M records), the framework demonstrated consistent improvements in authentication decisions, increasing 'Allow Login' actions and decreasing 'Deny Login' actions.
    • Explainable AI (SHAP) was utilized to enhance model transparency and interpretability, supporting informed decision-making.

    Conclusions:

    • The proposed hybrid RBA framework is accurate, adaptive, and effective for real-world applications.
    • The integration of machine learning, fuzzy logic, and optimization significantly improves account takeover detection and risk assessment.
    • The framework offers a promising solution for enhancing online security while maintaining a positive user experience.