基于机器学习的不平衡数据下个人信用评分预测模型研究  

Research on Personal Credit Rating Prediction Model Based on Machine Learning for Unbalanced Data

作　　者：费振华[1] FEl Zhenhua(Gansu Provincial Computing Center Lanzhou 730030,Gansu Province)

机构地区：[1]甘肃省计算中心,甘肃兰州730030

出　　处：《长江信息通信》2024年第4期112-114,共3页Changjiang Information & Communications

摘　　要：文章介绍了个人信用评分的基本概念,以及不平衡数据及其处理方法和机器学习算法在信用评分中的应用。然后,通过数据预处理,包括数据来源与特性、数据清洗与整理、数据不平衡分析、数据增强方法和效果评估,为后续模型构建提供基础。最后,使用实际数据集进行模型训练和测试,并评估模型的性能。实验结果表明,基于机器学习的不平衡数据下个人信用评分预测模型能够有效地预测个人信用风险,对于金融机构的风险管理和信贷决策具有重要意义。This article aims to study a personal credit score prediction model based on machine learning under imbalanced data.Firstly,the basic concept of personal credit scoring was introduced,as well as the application of imbalanced data and its proccssing methods,as well as machine learning algorithms in credit scoring.Then,through data preproccssing,including data sources and characteristics,data cleaning and organization,data imbalance analysis,data augmentation methods and effectiveness evaluation,a foundation is provided for subsequent model construction.Finally,use actual datasets for model training and testing,and evaluate the performance of the model.The experimental results show that the personal credit score prediction model based on machine learning under imbalanced data can effectively predict personal credit risk,which is of great significance for risk management and credit decision-making of financial institutions.

关 键 词：个人信用评分 不平衡数据 机器学习 数据预处理 模型研究 

分 类 号：TP309[自动化与计算机技术—计算机系统结构]