过采样数据建立模型中，所有模型AUC均大于0.800，其中，XGBoo

In the model building of oversampled data, the AUC of all models is greater than 0.800. Among them, the performance of the XGBoost model is better than other models, the accuracy rate: 85.1%, the precision: 0.810, the recall rate: 0.909, the F1 score: 0.857, and the AUC value is high Based on the AUC of the LR model, it shows that the diagnostic efficiency of the machine learning algorithm is higher than that of the traditional LR model, and it has good predictive performance.

In the oversampling data establishment model, the AUC of all models is greater than 0.800. Among them, the XGBoost model is superior to other models in terms of accuracy: 85.1%, accuracy: 0.810, recall rate: 0.909, F1 score: 0.857, AUC value and higher than the AUC of the LR model, which indicates that the diagnostic efficiency of machine learning algorithm is higher than that of the traditional LR model and has good prediction performance.

In the models established by over-sampled data, the AUC of all models is greater than 0.800, among which the performance of XGBoost model is better than other models, with accuracy of 85.1%, precision of 0.810, recall of 0.909, F1 score of 0.857 and AUC value higher than that of LR model, which shows that the diagnosis efficiency of machine learning algorithm is higher than that of traditional LR model and has good prediction performance.