基于大数据的贷款客户信息修复与智能决策系统设计  

DESIGN OF LOAN CUSTOMER INFORMATION RESTORATION AND INTELLIGENT DECISION SYSTEM BASED ON BIG DATA

作　　者：游鸽 许智超 肖建华 You Ge;Xu Zhichao;Xiao Jianhua(Nanfang College of Sun Yat-Sen University,Guangzhou 510970,Guangdong,China;School of Management,Jinan University,Guangzhou 510632,Guangdong,China;School of Public Administration and Emergency Management,Jinan University,Guangzhou 510632,Guangdong,China)

机构地区：[1]中山大学南方学院,广东广州510970 [2]暨南大学管理学院,广东广州510632 [3]暨南大学公共管理学院/应急管理学院,广东广州510632

出　　处：《计算机应用与软件》2023年第11期16-23,92,共9页Computer Applications and Software

摘　　要：为解决逾期高阶阶段出现的客户失联、资产情况无法评估、催收策略选择等问题,结合大数据技术构建贷款客户信息修复与智能决策系统。系统主要包括以下三个模块。数据集成模块:通过集成异构、碎片化的社交数据,形成结构化的文本数据集;数据挖掘模块:对文本数据集中的联系方式、地址、资产等信息进行识别和抽取,并基于词频证据对客户的联系方式、地址、资产等信息进行推理与判别;智能决策模块:基于if-then规则对贷款客户各特征属性下的催收策略分别打分,并利用多属性决策模型选择合适的催收策略。研究为贷款客户信息修复与智能决策系统的构建提供了全套的技术解决方案,具有一定的理论意义与实践意义。In order to solve the problems of customer out of communication,asset situation assessment failure,collection strategy selection and other issues in the late stage,big data technology was used to develop a system for loan customer information repair and intelligent collection decision.The system mainly included the following three modules.Data integration module integrated heterogeneous and fragmented social data to form a structured text data set.Data mining module identified and extracted contact information,address,assets and other information in the text data set,and reasoned and discriminated contact information,address,assets and other information of loan customers based on word frequency evidence.Intelligent decision module scored the collection strategies under each characteristic attribute of the loan customers based on if-then rule,and used the multi-attribute decision model to select the appropriate collection strategy.This research provided a complete set of technical solutions for the construction of loan customer information repair and intelligent decision system,which had certain theoretical and practical significance.

关 键 词：大数据 贷款客户 信息修复 智能决策 系统 

分 类 号：TP182[自动化与计算机技术—控制理论与控制工程] TP3[自动化与计算机技术—控制科学与工程]