联邦学习框架下基于Logistic回归的金融欺诈检测模型优化研究

谭纪新; 王洪

doi:10.11959/j.issn.2096-0271.2026043

您当前的位置：

首页 >

文章列表页 >

联邦学习框架下基于Logistic回归的金融欺诈检测模型优化研究

更新时间：2026-03-27

- 联邦学习框架下基于Logistic回归的金融欺诈检测模型优化研究
- Research on optimization of financial fraud detection model based on logistic regression under federated learning framework
- 大数据 2026年页码：1-15
- 作者机构：
  
  中南大学数学与统计学院，湖南长沙 410083
- 作者简介：
  
  [ "谭纪新（2003- ），女，中南大学硕士生，主要研究方向为分布式学习。" ]
  [ "王洪（1977- ），男，副教授，博士后，主要研究方向为分布式学习、高维数据分析、大数据分析、生存分析。" ]
- 基金信息：
  
  湖南省社科基金资助项目(23YBA009)
- DOI：10.11959/j.issn.2096-0271.2026043
  中图分类号： TP181
- 网络首发：2026-03-27，
- 稿件说明：
移动端阅览
谭纪新.联邦学习框架下基于Logistic回归的金融欺诈检测模型优化研究[J].大数据,

Tan Jixin.Research on optimization of financial fraud detection model based on logistic regression under federated learning framework[J].BIG DATA RESEARCH,
谭纪新.联邦学习框架下基于Logistic回归的金融欺诈检测模型优化研究[J].大数据, DOI：10.11959/j.issn.2096-0271.2026043.

Tan Jixin.Research on optimization of financial fraud detection model based on logistic regression under federated learning framework[J].BIG DATA RESEARCH, DOI：10.11959/j.issn.2096-0271.2026043.

摘要

Logistic回归作为金融风控领域广泛采用的分类算法，在传统集中式训练模式下面临数据隐私泄露风险与处理效率瓶颈。为应对这些挑战，该研究构建基于Logistic回归的联邦学习框架，创新性提出集成SMOTE、SOS与HT的优化模型，在保持联邦学习隐私保护优势的前提下，实现模型性能与效率的双重提升。实验表明，该研究提出的预测模型在不平衡数据处理中展现出更优越的预测性能，该模型受试者工作特征曲线下面积（AUC）值达0.9971，较传统联邦Logistic模型提升约7%，模型改进前后的召回率从32.25%提升至97.57%，且降低了总通信开销，实现了识别能力与通信效率的协同优化。

Abstract

Logistic regression

as a widely adopted classification algorithm in the financial risk control field

faces the risks of data privacy leakage and bottlenecks in processing efficiency under the traditional centralized training mode. To address these challenges

this study constructs a federated learning framework based on logistic regression

innovatively proposes an optimized model that integrates SMOTE

SOS

and HT. While maintaining the privacy protection advantages of federated learning

the framework achieves dual improvements in model performance and efficiency. Experimental results show that the proposed predictive model exhibits superior predictive performance in handling imbalanced data: the model's AUC value reaches 0.9971

which is 7% higher than that of the traditional federated

logistic model; the recall rate increases from 32.25% to 97.57%. Additionally

the model reduces the total communication overhead

realizing the coordinated optimization of recognition capability and communication efficiency.

关键词

Keywords

references

朱建明 , 张沁楠 , 高胜 , 等 . 基于区块链的隐私保护可信联邦学习模型 [J ] . 计算机学报 , 2021 , 44 ( 12 ): 2464 - 2484 .

Zhu J M , Zhang Q N , Gao S , et al . Privacy preserving and trustworthy federated learning model based on blockchain [J ] . Chinese Journal of Computers , 2021 , 44 ( 12 ): 2464 - 2484 .

Konečný J , McMahan B , Ramage D . Federated optimization: distributed optimization beyond the datacenter [PP ] . arXiv preprint , 2015 , arXiv: 1511.03575 .

Yang Q , Liu Y , Chen T J , et al . Federated machine learning: concept and applications [J ] . ACM Transactions on Intelligent Systems and Technology , 2019 , 10 ( 2 ): 1 - 19 .

王丹阳 , 梁伟红 , 李玉萍 , 等 . 面向不平衡数据的木薯叶部病害图像识别方法 [J ] . 中国农机化学报 , 2025 , 46 ( 3 ): 101 - 107, 123 .

Wang D Y , Liang W H , Li Y P , et al . Cassava leaf disease image recognition method for imbalanced data [J ] . Journal of Chinese Agricultural Mechanization , 2025 , 46 ( 3 ): 101 - 107, 123 .

王璐 , 程晓荣 . 改进不平衡数据中的集成学习算法 [J ] . 计算机与数字工程 , 2025 , 53 ( 1 ): 26 - 30 .

Wang L , Cheng X R . Improve ensemble learning algorithms in unbalanced data [J ] . Computer and Digital Engineering , 2025 , 53 ( 1 ): 26 - 30 .

Yang R , Jia S , Wang H L , et al . Dilated dynamic supervised contrastive learning framework for fault diagnosis under imbalanced dataset conditions [J ] . Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science , 2025 , 239 ( 7 ): 2626 - 2636 .

Salah Fathi K , Barakat S , Rezk A . An effective SQL injection detection model using LSTM for imbalanced datasets [J ] . Computers & Security , 2025 , 153 : 104391 .

汤殿华 , 曹云飞 , 黄云帆 , 等 . 基于CKKS同态加密的纵向联邦逻辑回归训练方案 [J ] . 通信技术 , 2024 , 57 ( 11 ): 1173 - 1180 .

Tang D H , Cao Y F , Huang Y F , et al . A vertical federated logistic regression training scheme based on CKKS homomorphic encryption [J ] . Communications Technology , 2024 , 57 ( 11 ): 1173 - 1180 .

雷大江 , 杜萌 , 李智星 , 等 . 稀疏多元逻辑回归问题优化算法研究 [J ] . 重庆邮电大学学报(自然科学版) , 2019 , 31 ( 3 ): 354 - 366 .

Lei D J , Du M , Li Z X , et al . Research on optimization algorithm for sparse multinomial logistic regression problem [J ] . Journal of Chongqing University of Posts and Telecommunications (Natural Science Edition) , 2019 , 31 ( 3 ): 354 - 366 .

Ding W Y , Ng M K , Zhang W X . A generalized alternating direction implicit method for consensus optimization: application to distributed sparse logistic regression [J ] . Journal of Global Optimization , 2024 , 90 ( 3 ): 727 - 753 .

王俊豪 . 联邦学习场景下的贡献评估和模型优化 [D ] . 中国科学技术大学 , 2024 .

Wang J H . Contribution Evaluation and Model Optimization in Federated Learning Scenarios [D ] . University of Science and Technology of China , 2024 .

王晓霞 , 李雷孝 , 林浩 . SMOTE类算法研究综述 [J ] . 计算机科学与探索 , 2024 , 18 ( 5 ): 1135 - 1159 .

Wang X X , Li L X , Lin H . Survey of research on SMOTE type algorithms [J ] . Journal of Frontiers of Computer Science & Technology , 2024 , 18 ( 5 ): 1135 - 1159 .

杜睿山 , 黄玉朋 , 付晓飞 , 等 . 基于SMOTE和XGBoost的天然气水合物与天然气储层识别 [J ] . 特种油气藏 , 2024 , 31 ( 5 ): 11 - 19 .

Du R S , Huang Y P , Fu X F , et al . Identification of natural gas hydrates and natural gas reservoirs based on SMOTE and XGBoost [J ] . Special Oil & Gas Reservoirs , 2024 , 31 ( 5 ): 11 - 19 .

王健宗 , 孔令炜 , 黄章成 , 等 . 联邦学习隐私保护研究进展 [J ] . 大数据 , 2021 , 7 ( 3 ): 130 - 149 .

Wang J Z , Kong L W , Huang Z C , et al . Research advances on privacy protection of federated learning [J ] . Big Data Research , 2021 , 7 ( 3 ): 130 - 149 .

张纪林 , 顾小卫 , 张亦钊 , 等 . 跨域数据授权运营研究及应用 [J ] . 大数据 , 2023 , 9 ( 4 ): 83 - 97 .

Zhang J L , Gu X W , Zhang Y Z , et al . Research and application of cross-domain data authorization and operation [J ] . Big Data Research , 2023 , 9 ( 4 ): 83 - 97 .

Qi X , Yang H B , Meng F , et al . Distributed photovoltaic short-term power predicition method based on horizontal federal learning [J ] . Acta Energiae Solaris Sinica , 2025 , 46 ( 1 ): 686 - 695 .

吴文泰 , 吴应良 , 林伟伟 , 等 . 横向联邦学习: 研究现状、系统应用与挑战 [J ] . 计算机学报 , 2025 , 48 ( 1 ): 35 - 67 .

Wu W T , Wu Y L , Lin W W , et al . Horizontal federated learning: research status, system applications and open challenges [J ] . Chinese Journal of Computers , 2025 , 48 ( 1 ): 35 - 67 .

孙爽 , 李晓会 , 刘妍 , 等 . 不同场景的联邦学习安全与隐私保护研究综述 [J ] . 计算机应用研究 , 2021 , 38 ( 12 ): 3527 - 3534 .

Sun S , Li X H , Liu Y , et al . Survey on security and privacy protection in different scenarios of federated learning [J ] . Application Research of Computers , 2021 , 38 ( 12 ): 3527 - 3534 .

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

暂无数据