IDEAS home Printed from https://ideas.repec.org/a/gvu/iroism/v7y2025i2id1052.html

Smart and Explainable Credit Card Fraud Detection Using XGBoost and SHAP

Author

Listed:
  • Selvam S.
  • Sughasiny M.

Abstract

Credit card fraud is a nagging problem in the world of credit transactions, which significantly leads to massive economic losses, and undermines users' confidence. Conventional fraud detection mechanisms are typically not adaptive, nor interpretable, thus being unsuitable for emerging fraud patterns and financial environments driven by compliance. In this paper, we introduce a smart and explainable credit card fraud detection system, with “smart” being a keyword to indicate an adaptive, modular, and tunable model architecture specialized for imbalanced data, and “explainable” for providing a transparent and feature-level explanation for any decision made by the model, utilizing the SHAP (SHapley Additive exPlanations) technique. The model we implemented is composed of these two libraries: the method decides to use XGBoost as a classifier and takes Random Forest as a benchmark. The two models are trained and evaluated for performance on the imbalanced Kaggle Credit Card Fraud Detection dataset, using stratified 5-fold cross-validation and grid search for hyperparameter selection. The final XGBoost model is better able to distinguish between classes, with 92.1% precision and 87.3% recall. SHAP is integrated into the prediction pipeline as a means of creating instance-level explanations to achieve post hoc analysis and meet GDPR and PCI DSS compliance. These interpretations and predictions are supplied and protected via role-based access control and encryption for audit. Experimental results show the model’s power to accurately detect rare fraud examples in a transparent and operationally robust way. This work addresses the trade-off between prediction performance and interpretability, and enables safe, real-time fraud detection in contemporary financial institutions. It also provides a deployable design that satisfies regulatory requirements and an effective analyst workflow, making it applicable for a production-based financial security system.

Suggested Citation

Handle: RePEc:gvu:iroism:v:7:y:2025:i:2:id:1052
DOI: 10.36548/jismac.2025.2.004
as

Download full text from publisher

File URL: https://irojournals.com/iroismac/article/view/1052
File Function: Abstract page
Download Restriction: no

File URL: https://irojournals.com/iroismac/article/download/1052/938
File Function: Full text
Download Restriction: no

File URL: https://libkey.io/10.36548/jismac.2025.2.004?utm_source=ideas
LibKey link: if access is restricted and if your library uses this service, LibKey will redirect you to where you can use your library subscription to access this item
---><---

More about this item

Keywords

;
;
;
;
;
;
;
;
;
;

Statistics

Access and download statistics

Corrections

All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:gvu:iroism:v:7:y:2025:i:2:id:1052. See general information about how to correct material in RePEc.

If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

We have no bibliographic references for this item. You can help adding them by using this form .

If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: Journal Manager (email available below). General contact details of provider: https://irojournals.com/iroismac .

Please note that corrections may take a couple of weeks to filter through the various RePEc services.

IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.