从系统论的角度总结了个人信用评分发展的前沿问题,从数据预处理、指标体系筛选、以及模型设计三个方面对个人信用评分关键技术的最新研究成果进行了细致分类和综合比较,从而指出个人信用评分研究中存在的难点以及未来发展方向。
Most early researches on credit scoring were restricted to establish and improve an scoring model. Researchers are now regarding credit scoring as an integrated system involving da- ta preprocessing, feature selection and model design. Academics are applying latest research a- chievements in statistics, data mining and machine learning to the whole process of credit sco- ring, making studies on credit scoring gain an overall and rapid development.