基于机器学习的个人信用模型实证分析

Empirical A nalysis of Personal Credit Model Based on Machine Learning

信用是金融市场发展的基础.个人信用评价不仅关系到金融效率与金融风险,更涉及到全民素质的提升.基于P2P平台的调查数据,运用多种机器学习算法对个人数据进行处理并比较,发现机器学习的个人信用模型对处理小样本数据同样具有优势,其中随机森林捕捉正常用户准确率高,支持向量机(SVM)捕捉违约用户准确率高.

Credit is the foundation of the development of financial market.Personal credit evaluation is not only related to financial eficiency and financial risk,but also to the improvement of the quality of the whole population..Based on the survey data of P2P platform,this paper uses a variety of machine learning algorithms to process and compare personal data,and finds that the personal credit model of machine learning has the same advantages in dealing with small sample data.Random forest has high accuracy in capturing normal users and SVM has high accuracy in capturing default users.