Academic Journal of Business & Management, 2022, 4(4); doi: 10.25236/AJBM.2022.040402.
Chuyu Feng1, Ling Gu2
1School of International Business Administration, South China Normal University, Guangzhou, Guangdong, 510000, China
2School of Economics and Management, Southwest Jiaotong University, Chengdu, Sichuan, 610000, China
With the development of large-scale data technology, the Internet finance industry is developing rapidly. Research on personal credit risk assessment models is conducive to strengthening risk control and management and improving the efficiency of accurate services by financial institutions. Based on this, this paper first considers the amount of information, independence and relevance of the data, and uses four measurement to measure the combination of credit impact factors such as basic personal information, basic credit information and credit behaviour information. Four machine learning methods are then used to assess individual credit risk and give a comprehensive comparison of the degree of importance of credit risk impact factors on individual credit risk. The empirical results show that the model is accurate and stable, and can well reflect the degree of influence of individual credit characteristics on credit risk. Finally, providing systematic construction suggestions of risk management in banks and other financial institutions.
Personal Credit Risk, Impact Factors, Risk Assessment, Machine Learning
Chuyu Feng, Ling Gu. A Study on the Measurement and Evaluation of Personal Credit Risk Impact Factors Based on Machine Learning. Academic Journal of Business & Management (2022) Vol. 4, Issue 4: 7-10. https://doi.org/10.25236/AJBM.2022.040402.
[1] Cai Wenwen, Luo Yonghao, Zhang Guanxiang, Zhong Huiling Personal credit risk assessment model and empirical analysis based on the integration of gbdt and logistic regression [J]. Management modernization, 2017, 37 (02): 1-4.
[2] She Chaobing, Application of logistic regression in bank personal credit risk assessment [J]. Technology and innovation, 2018 (19): 113-114 + 118-119.