Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

Applying explainable artificial intelligence to interpret supervised ensemble learning models for robust credit card

Shimaa S Awad¹, Alyaa A Hamza², Mohamed A Sobh^3,2

¹Computer Engineering & Systems Department, Faculty of Engineering, Ain Shams University, Cairo, Egypt. 2101391@eng.asu.edu.eg.

Scientific Reports

|May 15, 2026

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

Effect of probiotic and synbiotic administration on the nutritional status of hemodialysis patients: a randomized-controlled trial.

International urology and nephrology·2026

Same author

Feature importance guided autoencoder for dimensionality reduction in intrusion detection systems.

Scientific reports·2026

Same author

Efficient feature ranked hybrid framework for android Iot malware detection.

Scientific reports·2026

Same author

Optimized CatBoost machine learning (OCML) for DDoS detection in cloud virtual machines with time-series and adversarial robustness.

Scientific reports·2026

Same author

HSAS-MD Analyzer: A Hybrid Security Analysis System Using Model-Checking Technique and Deep Learning for Malware Detection in IoT Apps.

Sensors (Basel, Switzerland)·2022

Same author

Effect of Everolimus versus Bone Marrow-Derived Stem Cells on Glomerular Injury in a Rat Model of Glomerulonephritis: A Preventive, Predictive and Personalized Implication.

International journal of molecular sciences·2022

This summary is machine-generated.

This study evaluates machine learning models for credit card fraud detection, finding that tree-based ensembles like XGBoost offer high accuracy and interpretability using Explainable Artificial Intelligence (XAI) techniques.

Area of Science:

Computer Science
Data Science
Financial Technology

Background:

Digital financial transactions are rapidly increasing, necessitating advanced fraud detection methods.
Complex machine learning models offer high accuracy but often lack transparency, hindering trust and accountability.
Limited research exists on the interpretability of high-performing credit card fraud detection models.

Purpose of the Study:

To assess the predictive performance and explainability of four supervised learning algorithms for credit card fraud detection.
To investigate the applicability of Explainable Artificial Intelligence (XAI) techniques in conjunction with these models.
To bridge the research gap concerning the interpretability of accurate fraud detection systems.

Main Methods:

Keywords:

Explainable Artificial Intelligence (XAI)Financial fraud Fraud detection Machine learning Model interpretability Supervised learning

Related Experiment Videos

Evaluated Logistic Regression, Random Forest, XGBoost, and LightGBM on three public credit card transaction datasets.

Employed performance metrics including accuracy, precision, recall, F1-score, and ROC-AUC.

Utilized the SHAP (SHapley Additive exPlanations) framework to interpret model predictions and identify feature importance.

Main Results:

Tree-based ensemble models, particularly XGBoost, demonstrated superior predictive performance.
Linear models also showed improvements over baseline methods.
SHAP analysis successfully identified key features and provided insights into complex model outputs.

Conclusions:

XGBoost and other tree-based ensembles provide a strong balance of predictive accuracy and interpretability for credit card fraud detection.
XAI techniques like SHAP are crucial for understanding and trusting high-performing fraud detection models.
Findings support the development of more transparent and accountable financial security systems.