Skip to main content
. 2022 Aug 26;12:897596. doi: 10.3389/fonc.2022.897596

Table 2.

Performance of the XGBoost model.

Training dataset (N = 469) Test dataset (N = 118)
BA 84.89% 85.21%
F-score 73.36% 76.06%
MCC 63.68% 66.56%
Precision 63.10% 69.23%
Recall 87.60% 84.38%
AUC 91.53% 90.88%
RMSE 0.4052 0.3796
R2 0.1424 0.2711

XGBoost, eXtreme Gradient Boosting; BA, balanced accuracy; MCC, Matthew’s correlation coefficient; AUC, area under the curve; RMSE, root mean square error; R2, coefficient of determination.