article · FUDMA Journal of Sciences
Credit card fraud poses serious financial challenges to institutions, largely because transaction data contains far fewer fraudulent events than legitimate ones. To overcome this severe class imbalance, a new machine learning framework applies SMOTE-Tomek Links to synthesise fraud samples while removing borderline majority-class records. Key predictors are selected using Recursive Feature Elimination, and model parameters are tuned automatically via RandomizedSearchCV. The framework combines Random Forest and XGBoost base classifiers through a stacked Logistic Regression ensemble. When evaluated across multiple performance metrics, the stacked approach outperforms its individual component models. It records an Area Under the Precision-Recall Curve score of 0.8207 and an F1-score of 0.93, restricting misclassifications to 20 false negatives and five false positives. This structure suppresses financial losses from fraud whilst preventing genuine customer transactions from being incorrectly declined.
Detecting payment card fraud accurately is difficult because fraudulent transactions are rare compared to legitimate ones. Poor detection models either miss real fraud or mistakenly block valid purchases, frustrating customers. By balancing data and combining predictive algorithms, this framework demonstrates how financial institutions can significantly reduce monetary losses from cybercrime without disrupting normal shopping experiences for cardholders.
The framework is designed for banks and payment processing companies seeking operational tools to identify credit card fraud. By maintaining low false positives and negatives, it targets automated transaction monitoring systems. As an applied and tested model validated on transaction data, the technology represents late-stage research that could be integrated into existing financial fraud detection pipelines.
AI-generated from the published abstract. Always read the original work before citing.
Credit card fraud is a menace to financial institutions, but detection is compromised by highly imbalanced transaction datasets. This study proposes an advanced machine learning framework optimized for fraud detection. To address the issue of data imbalance, SMOTE-Tomek Links is applied to synthetically generate minority fraud cases while removing noisy, overlapping majority-class instances. Recursive Feature Elimination (RFE) is deployed to identify the optimal features, and RandomizedSearchCV automates hyperparameter optimization. The study introduces a Stacked Logistic Regression ensemble to combine the predictive capacity of optimized Random Forest and XGBoost base classifiers. The model’s effectiveness is assessed using seven evaluation methods: accuracy, recall, precision, confusion matrix, F1-score, Receiver Operating Characteristic Area Under the Curve (ROC-AUC) score and the Area Under the Precision-Recall Curve (AUC-PR) score. Findings reveal that the proposed stacked model performance surpasses both individual base models. While achieving deceptively high baseline accuracy across all models, the stacked ensemble delivers a superior AUC-PR score of 0.8207 and an F1-score of 0.93. This minimizes the confusion matrix misclassifications to just 20 False Negatives and 5 False Positives. The framework provides a cost-optimized operational engine that aggressively mitigates bank fraud losses while successfully shielding legitimate cardholders from accidental checkout declines.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.33003/fjs-2026-1014-5398
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
