大数据背景下商业银行个人信用评估体系构建

Construction of Personal Credit Evaluation System of Commercial Banks under the Background of Big Data

构建指标体系是个人信贷风险识别的基础性工作。大数据背景下,商业银行信用评估数据源更加广泛。因此,基于传统征信系统和互联网平台数据,从用户基本静态信息、用户财务能力、用户信贷及抵押历史、用户交易特征、用户行为偏好、用户社交关系等六个维度构建商业银行个人信用评估指标体系,依据数据结构特征,采用主成分分析和机器学习的组合方法对指标体系降维压缩。该体系弥补了传统信用评估中数据来源单一的缺陷,并降低了指标数据处理的复杂度。

The construction of index system is the basic problem of personal credit risk identification.Under the background of big data,credit evaluation data sources of commercial banks are more extensive.Based on the data of traditional credit investigation system and Inter-net platform,the personal credit evaluation index system of commercial banks is constructed from the six dimensions of basic static informa-tion of users,financial ability of users,credit and mortgage history of users,transaction characteristics of users,behavioral preference of users and social relations of users.According to the characteristics of data structure,principal component analysis and machine learning were used to reduce dimension of index system.It makes up for the defect of single data source in traditional credit evaluation and reduces the complex-ity of index data processing.