FinGraphFL: Financial Graph-Based Federated Learning for Enhanced Credit Card Fraud Detection

In the field of credit card fraud detection, traditional methods often struggle due to their reliance on complex manual feature engineering or their inability to adapt to rapidly changing fraud patterns. This paper introduces an innovative approach called FinGraphFL, which merges graph-based learnin...

Full description

Saved in:
Bibliographic Details
Main Authors: Zhenyu Xia, Suvash C. Saha
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Mathematics
Subjects:
Online Access:https://www.mdpi.com/2227-7390/13/9/1396
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850155740090597376
author Zhenyu Xia
Suvash C. Saha
author_facet Zhenyu Xia
Suvash C. Saha
author_sort Zhenyu Xia
collection DOAJ
description In the field of credit card fraud detection, traditional methods often struggle due to their reliance on complex manual feature engineering or their inability to adapt to rapidly changing fraud patterns. This paper introduces an innovative approach called FinGraphFL, which merges graph-based learning with the principles of federated learning and improves security through differential privacy. FinGraphFL utilizes Graph Attention Networks to analyze dynamic relationships between daily credit card transaction records, enhancing its ability to detect fraudulent activities. With the addition of differential privacy, the model allows multiple financial institutions to collaborate to refine the detection model without sharing sensitive data, thus improving adaptability and accuracy. The results are tested in two public datasets that show that FinGraphFL achieves accuracy rates of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>0.9780</mn></mrow></semantics></math></inline-formula> and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>0.9839</mn></mrow></semantics></math></inline-formula>, significantly outperforming traditional methods. Building on these results, FinGraphFL sets the stage for future advances in real-time learning and global financial collaboration, ensuring simultaneous progress in security and privacy protections.
format Article
id doaj-art-e71fa7eb56da4cd0aae51ea27e7d75ad
institution OA Journals
issn 2227-7390
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Mathematics
spelling doaj-art-e71fa7eb56da4cd0aae51ea27e7d75ad2025-08-20T02:24:47ZengMDPI AGMathematics2227-73902025-04-01139139610.3390/math13091396FinGraphFL: Financial Graph-Based Federated Learning for Enhanced Credit Card Fraud DetectionZhenyu Xia0Suvash C. Saha1School of Mechanical and Mechatronic Engineering, Faculty of Engineering and Information Technology, University of Technology Sydney, Sydney, NSW 2007, AustraliaSchool of Mechanical and Mechatronic Engineering, Faculty of Engineering and Information Technology, University of Technology Sydney, Sydney, NSW 2007, AustraliaIn the field of credit card fraud detection, traditional methods often struggle due to their reliance on complex manual feature engineering or their inability to adapt to rapidly changing fraud patterns. This paper introduces an innovative approach called FinGraphFL, which merges graph-based learning with the principles of federated learning and improves security through differential privacy. FinGraphFL utilizes Graph Attention Networks to analyze dynamic relationships between daily credit card transaction records, enhancing its ability to detect fraudulent activities. With the addition of differential privacy, the model allows multiple financial institutions to collaborate to refine the detection model without sharing sensitive data, thus improving adaptability and accuracy. The results are tested in two public datasets that show that FinGraphFL achieves accuracy rates of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>0.9780</mn></mrow></semantics></math></inline-formula> and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>0.9839</mn></mrow></semantics></math></inline-formula>, significantly outperforming traditional methods. Building on these results, FinGraphFL sets the stage for future advances in real-time learning and global financial collaboration, ensuring simultaneous progress in security and privacy protections.https://www.mdpi.com/2227-7390/13/9/1396privacy preservingfederated learninglocal federalizationdifferential privacy
spellingShingle Zhenyu Xia
Suvash C. Saha
FinGraphFL: Financial Graph-Based Federated Learning for Enhanced Credit Card Fraud Detection
Mathematics
privacy preserving
federated learning
local federalization
differential privacy
title FinGraphFL: Financial Graph-Based Federated Learning for Enhanced Credit Card Fraud Detection
title_full FinGraphFL: Financial Graph-Based Federated Learning for Enhanced Credit Card Fraud Detection
title_fullStr FinGraphFL: Financial Graph-Based Federated Learning for Enhanced Credit Card Fraud Detection
title_full_unstemmed FinGraphFL: Financial Graph-Based Federated Learning for Enhanced Credit Card Fraud Detection
title_short FinGraphFL: Financial Graph-Based Federated Learning for Enhanced Credit Card Fraud Detection
title_sort fingraphfl financial graph based federated learning for enhanced credit card fraud detection
topic privacy preserving
federated learning
local federalization
differential privacy
url https://www.mdpi.com/2227-7390/13/9/1396
work_keys_str_mv AT zhenyuxia fingraphflfinancialgraphbasedfederatedlearningforenhancedcreditcardfrauddetection
AT suvashcsaha fingraphflfinancialgraphbasedfederatedlearningforenhancedcreditcardfrauddetection