With the rapid economic development of China, the importance of credit consumption methods in China's economy and people's daily lives has become increasingly prominent. Based on the GiveMeSomeCredit dataset, this paper constructs an XGBoost model to conduct in-depth predictive analysis on credit default probability issues. This paper first performs meticulous data cleaning and missing value handling on the dataset, and divides the dataset to prepare for model training. Subsequently, the XGBoost model is constructed and trained, and parameter optimization is further carried out during this process. Finally, the model's performance is evaluated using key evaluation indicators such as Accuracy, Precision, Recall, and AUC, and it is compared with the random forest model and logistic regression model. The results show that the XGBoost model performs better. It can be seen that the XGBoost model has high application value in credit default probability prediction.

Hong Rao, Chenhao Wei. Credit Default Probability Prediction Model Based on XGBoost Algorithm. Academic Journal of Computing & Information Science (2024), Vol. 7, Issue 10: 60-66. https://doi.org/10.25236/AJCIS.2024.071009.

[1] Wang Jianhai. Design and Implementation of Credit Default Prediction Model Based on Combined Feature Selection and LightGBM [D]. Southeast University, 2022.

[2] Zheng Cheng. Research on Bond Default Prediction Based on CNN-SVM Model [D]. Zhejiang University of Finance and Economics, 2024.

[3] Ni Xu. Research on Credit Evaluation of New Agricultural Business Entities in China [D]. Chinese Academy of Agricultural Sciences, 2019.

[4] Zhou Qing'an. Research on Personal Online Loan Credit Evaluation Based on Genetic XGBoost Model [D]. Jiangxi University of Finance and Economics, 2020.

[5] Zhou Rongxi, Peng Hang, Li Xinyu, et al. Credit Bond Default Prediction Model Based on XGBoost Algorithm [J]. Bonds, 2019.