{"title":"A novel communication-efficient heterogeneous federated positive and unlabeled learning method for credit scoring","authors":"Yongqin Qiu , Yuanxing Chen , Kan Fang , Kuangnan Fang","doi":"10.1016/j.cor.2025.106982","DOIUrl":null,"url":null,"abstract":"<div><div>Customer records include only customers in default (positive samples) and rejected customers (unlabeled samples), or positive and unlabeled (PU) data, which is a common scenario in emerging financial institutions. However, building credit scoring models using multiple small sample PU datasets with high dimensionality poses significant challenges, especially in light of the privacy constraints associated with transferring raw data. To tackle these challenges, this paper introduces a novel methodology called heterogeneous federated PU learning. This approach utilizes a fused penalty function to automatically divide coefficients into multiple clusters, while an efficient proximal gradient descent algorithm is introduced for model training, relying solely on gradients from local servers. Theoretical analysis establishes the oracle property of our proposed estimator. The simulation results show that, in terms of variable selection, parameter estimation, and prediction performance, our method is close to the Oracle estimator and outperforms the other alternatives. Empirical results indicate that our method can improve prediction performance and facilitate the identification of heterogeneity across datasets. Moreover, the estimated clustering structures further reveal that provinces that are geographically closer exhibit greater similarity in credit risk. This implies that the proposed methodology can effectively assist nascent financial institutions in identifying differences in risk factors across datasets and enhancing predictive accuracy.</div></div>","PeriodicalId":10542,"journal":{"name":"Computers & Operations Research","volume":"177 ","pages":"Article 106982"},"PeriodicalIF":4.1000,"publicationDate":"2025-01-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Computers & Operations Research","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0305054825000103","RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS","Score":null,"Total":0}
引用次数: 0
Abstract
Customer records include only customers in default (positive samples) and rejected customers (unlabeled samples), or positive and unlabeled (PU) data, which is a common scenario in emerging financial institutions. However, building credit scoring models using multiple small sample PU datasets with high dimensionality poses significant challenges, especially in light of the privacy constraints associated with transferring raw data. To tackle these challenges, this paper introduces a novel methodology called heterogeneous federated PU learning. This approach utilizes a fused penalty function to automatically divide coefficients into multiple clusters, while an efficient proximal gradient descent algorithm is introduced for model training, relying solely on gradients from local servers. Theoretical analysis establishes the oracle property of our proposed estimator. The simulation results show that, in terms of variable selection, parameter estimation, and prediction performance, our method is close to the Oracle estimator and outperforms the other alternatives. Empirical results indicate that our method can improve prediction performance and facilitate the identification of heterogeneity across datasets. Moreover, the estimated clustering structures further reveal that provinces that are geographically closer exhibit greater similarity in credit risk. This implies that the proposed methodology can effectively assist nascent financial institutions in identifying differences in risk factors across datasets and enhancing predictive accuracy.
期刊介绍:
Operations research and computers meet in a large number of scientific fields, many of which are of vital current concern to our troubled society. These include, among others, ecology, transportation, safety, reliability, urban planning, economics, inventory control, investment strategy and logistics (including reverse logistics). Computers & Operations Research provides an international forum for the application of computers and operations research techniques to problems in these and related fields.