基于风险权值聚合联邦学习的金融欺诈检测模型

孙元林; 王楠; 张焘; 刘娟; 赵沁馨

doi:10.11959/j.issn.2096-109x.2025056

您当前的位置：

首页 >

文章列表页 >

基于风险权值聚合联邦学习的金融欺诈检测模型

专题：大模型数据安全与隐私保护 | 更新时间：2025-11-08

- 基于风险权值聚合联邦学习的金融欺诈检测模型
- Financial fraud detection model based on risk weight aggregation federated learning
- 网络与信息安全学报 2025年11卷第5期页码：88-100
- 作者机构：
  
  1.北京交通大学软件学院，北京 100044
  2.吉林大学符号计算与知识工程教育部重点实验室，吉林长春 130012
  3.北京交通大学网络空间安全学院，北京 100044
  4.南京大学软件学院，江苏南京 210008
- 作者简介：
  
  [ "孙元林（2000- ），男，山东东营人，北京交通大学硕士生，主要研究方向为异常检测、网络安全。" ]
  [ "王楠（1993- ），女，山东潍坊人，北京交通大学副教授，主要研究方向为网络安全、恶意攻击分析。" ]
  [ "张焘（1996- ），男，安徽马鞍山人，北京交通大学副教授，主要研究方向为网络安全、联邦学习。" ]
  [ "刘娟（1987- ），女，河北衡水人，北京交通大学高级工程师，主要研究方向为金融欺诈检测、工业异常检测、联邦学习。" ]
  [ "赵沁馨（2002- ），女，江苏扬州人，主要研究方向为网络安全、恶意攻击分析。" ]
- 基金信息：
  
  国家自然科学基金(62202042);中央高校基本科研业务费专项基金(2024JBMC031);航空科学基金(ASFC-2024Z0710M5002);先进密码技术与系统安全四川省重点实验室开放课题资助项目(SKLACSS-202312)
- DOI：10.11959/j.issn.2096-109x.2025056
  中图分类号： TP311.1
- 收稿：2025-01-21，
  
  修回：2025-05-07，
  
  纸质出版：2025-10-25
- 稿件说明：
移动端阅览
孙元林, 王楠, 张焘, 等. 基于风险权值聚合联邦学习的金融欺诈检测模型[J]. 网络与信息安全学报, 2025, 11(5): 88-100.

SUN Y L, WANG N, ZHANG T, et al. Financial fraud detection model based on risk weight aggregation federated learning[J]. Chinese Journal of Network and Information Security, 2025, 11(5): 88-100.
孙元林, 王楠, 张焘, 等. 基于风险权值聚合联邦学习的金融欺诈检测模型[J]. 网络与信息安全学报, 2025, 11(5): 88-100. DOI： 10.11959/j.issn.2096-109x.2025056.

SUN Y L, WANG N, ZHANG T, et al. Financial fraud detection model based on risk weight aggregation federated learning[J]. Chinese Journal of Network and Information Security, 2025, 11(5): 88-100. DOI： 10.11959/j.issn.2096-109x.2025056.

摘要

金融欺诈给金融机构和消费者带来了巨大的损失，如何能够快速、有效、准确地识别并阻止金融欺诈行为成为人们普遍关注的问题。机器学习、深度学习等技术由于能够有效地检测动态多变的欺诈模式，得到了广泛研究与关注。然而，金融数据的隐私性、保密性等特点，使得深度学习等数据驱动的智能分析技术在金融领域的实际应用中受到限制。因此，研究如何在保障数据隐私安全的条件下进行协同训练并建立有效的欺诈检测模型具有重要的价值。针对上述问题，提出风险权值聚合联邦学习（risk weight aggregation federated learning

FedRWA）算法进行多模型融合，在保证数据隐私的前提下构建中央异常检测模型，实现多源敏感数据模型融合学习，使中央联合模型能够精准获取各机构私有数据集的欺诈风险特征，以提升欺诈检测效果。为了验证所提方法的有效性，使用信用卡交易数据集开展了实验，实验结果表明，FedRWA算法的性能优于多种前沿的联邦学习方法，能够检测出更多更大额的欺诈样本，同时进一步提升了隐私数据的安全性。

Abstract

Financial fraud had caused significant losses to financial institutions and consumers

making the rapid

effective

and accurate identification and prevention of financial fraud a widely concerned issue. Machine learning and deep learning technologies had received extensive research attention due to their capability to effectively detect dynamic and changing fraud patterns. However

the privacy and confidentiality requirements of financial data were found to limit the practical application of data-driven intelligent analysis technologies

including deep learning

in the financial field. Therefore

research on how to conduct collaborative training and establish effective fraud detection models while ensuring data privacy and security was considered to be of great value. To address these challenges

a risk weight aggregation federated learning (FedRWA) method was proposed for multi-model fusion. A central anomaly detection model was constructed to achieve multi-source sensitive data model fusion learning while ensuring data privacy. This approach enabled the central joint model to accurately capture the fraud risk characteristics of each institution’s private dataset

thereby improving fraud detection effectiveness. To verify the effectiveness of the proposed method

experiments were conducted using a credit card transaction dataset. The experimental results demonstrate that the FedRWA method performs better than numerous cutting-edge federated learning methods

could detect more and larger fraud samples

and further enhances the security of private data.

关键词

Keywords

references

HILAL W , GADSDEN S A , YAWNEY J . Financial fraud: a review of anomaly detection techniques and recent advances [J ] . Expert Systems with Applications , 2022 , 193 : 116429 .

GIANINI G , FOSSI L G , MIO C , et al . Managing a pool of rules for credit card fraud detection by a game theory based approach [J ] . Future Generation Computer Systems , 2020 , 102 : 549 - 561 .

杨强 . AI与数据隐私保护：联邦学习的破解之道 [J ] . 信息安全研究 , 2019 , 5 ( 11 ): 961 - 965 .

YANG Q . AI and data privacy protection:the cracking method of federated learning [J ] . Information Security Research , 2019 , 5 ( 11 ): 961 - 965 .

MAKKI S , ASSAGHIR Z , TAHER Y , et al . An experimental study with imbalanced classification approaches for credit card fraud detection [J ] . IEEE Access , 2019 , 7 : 93010 - 93022 .

XIAO C , WANG S . An experimental study of class imbalance in federated learning [C ] // Proceedings of 2021 IEEE Symposium Series on Computational Intelligence (SSCI) . Piscataway : IEEE Press , 2021 : 1 - 7 .

MOTIE S , RAAHEMI B . Financial fraud detection using graph neural networks: a systematic review [J ] . Expert Systems with Applications , 2024 , 240 : 122156 .

ABDALLAH A , MAAROF M A , ZAINAL A . Fraud detection system: a survey [J ] . Journal of Network and Computer Applications , 2016 , 68 : 90 - 113 .

YANG G , LIU X , LI B . Anti-money laundering supervision by intelligent algorithm [J ] . Computers & Security , 2023 , 132 : 103344 .

SAHIN Y , BULKAN S , DUMAN E . A cost-sensitive decision tree approach for fraud detection [J ] . Expert Systems with Applications , 2013 , 40 ( 15 ): 5916 - 5923 .

SAHIN Y , DUMAN E . Detecting credit card fraud by ANN and logistic regression [C ] // Proceedings of 2011 International Symposium on Innovations in Intelligent Systems and Applications . Piscataway : IEEE Press , 2011 : 315 - 319 .

GYAMFI N K , ABDULAI J D . Bank fraud detection using support vector machine [C ] // Proceedings of 2018 IEEE 9th Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON) . Piscataway : IEEE Press , 2018 : 37 - 41 .

ALGHOFAILI Y , ALBATTAH A , RASSAM M A . A financial fraud detection model based on LSTM deep learning technique [J ] . Journal of Applied Security Research , 2020 , 15 ( 4 ): 498 - 516 .

MCMAHAN B , MOORE E , RAMAGE D , et al . Communication-efficient learning of deep networks from decentralized data [C ] // Proceedings of Artificial Intelligence and Statistics . 2017 : 1273 - 1282 .

LI T , SAHU A K , ZAHEER M , et al . Federated optimization in heterogeneous networks [J ] . Machine Learning and Systems , 2020 , 2 : 429 - 450 .

应作斌 , 方一晨 , 张怡文 . 动态聚合权重的隐私保护联邦学习框架 [J ] . 网络与信息安全学报 , 2022 , 8 ( 5 ): 56 - 65 .

YING Z B , FANG Y C , ZHANG Y W . Privacy-preserving federated learning framework with dynamic weight aggregation [J ] . Chinese Journal of Network and Information Security , 2022 , 8 ( 5 ): 56 - 65 .

余锋 , 林庆新 , 林晖 , 等 . 基于生成对抗网络的隐私增强联邦学习方案 [J ] . 网络与信息安全学报 , 2023 , 9 ( 3 ): 113 - 122 .

YU F , LIN Q X , LIN H , et al . Privacy-enhanced federated learning scheme based on generative adversarial networks [J ] . Chinese Journal of Network and Information Security , 2023 , 9 ( 3 ): 113 - 122 .

BONAWITZ K , EICHNER H , GRIESKAMP W , et al . Towards federated learning at scale: system design [J ] . Machine Learning and Systems , 2019 , 1 : 374 - 388 .

WANG J , LIU Q , LIANG H , et al . Tackling the objective inconsistency problem in heterogeneous federated optimization [J ] . Advances in Neural Information Processing Systems , 2020 , 33 : 7611 - 7623 .

LI Z , LIN T , SHANG X , et al . Revisiting weighted aggregation in federated learning with neural networks [C ] // Proceedings of International Conference on Machine Learning . 2023 : 19767 - 19788 .

DENG Y , LYU F , REN J , et al . Improving federated learning with quality-aware user incentive and auto-weighted model aggregation [J ] . IEEE Transactions on Parallel and Distributed Systems , 2022 , 33 ( 12 ): 4515 - 4529 .

张连福 , 谭作文 . 一种基于自适应加权的鲁棒联邦学习算法 [J ] . 计算机科学 , 2023 , 50 ( z1 ): 799 - 807 .

ZHANG L F , TAN Z W . Robust federated learning algorithm based on adaptive weighting [J ] . Computer Science , 2023 , 50 ( z1 ): 799 - 807 .

杨智凯 , 刘亚萍 , 张硕 , 等 . 联邦学习通信优化方法综述 [J ] . 网络与信息安全学报 , 2024 , 10 ( 6 ): 1 - 23 .

YANG Z K , LIU Y P , ZHANG S , et al . Review of communication optimization methods in federated learning [J ] . Chinese Journal of Network and Information Security , 2024 , 10 ( 6 ): 1 - 23 .

刘天瑞 , 王连海 , 王启正 , 等 . 算力网络环境下基于区块链的自适应联邦学习 [J ] . 网络与信息安全学报 , 2024 , 10 ( 3 ): 130 - 142 .

LIU T R , WANG L H , WANG Q Z , et al . Blockchain based adaptive federated learning in computational power network [J ] . Chinese Journal of Network and Information Security , 2024 , 10 ( 3 ): 130 - 142 .

YU F , RAWAT A S , MENON A , et al . Federated learning with only positive labels [C ] // Proceedings of International Conference on Machine Learning . 2020 : 10946 - 10956 .

KAIROUZ P , MCMAHAN H B , AVENT B , et al . Advances and open problems in federated learning [J ] . Foundations and Trends in Machine Learning , 2021 , 14 ( 1-2 ): 1 - 210 .

KUBAT M , HOLTE R C , MATWIN S . Machine learning for the detection of oil spills in satellite radar images [J ] . Machine Learning , 1998 , 30 ( 2 ): 195 - 215 .

LI Q , HE B , SONG D . Model-contrastive federated learning [C ] // Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition . 2021 : 10713 - 10722 .

LI X , JIANG M , ZHANG X , et al . FedBN: federated learning on non-IID features via local batch normalization [J ] . arXiv preprint arXiv: 2102.07623 , 2021 .

KIM S M , SHIN W , JANG S , et al . FedRN: exploiting k-reliable neighbors towards robust federated learning [C ] // Proceedings of the 31st ACM International Conference on Information & Knowledge Management . 2022 : 972 - 981 .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

DP-FedDRC：缓解维度坍塌的差分隐私个性化联邦学习

基于动态聚类和信誉评估的抗攻击联邦框架

FL-IoV：基于联邦学习的综合车联网入侵检测方法

FLWD:基于联邦学习的Webshell检测方法

Robust-PFedproto:基于个性化层的鲁棒联邦原型学习方法