Skip to main content
. 2022 Sep 27;5:979465. doi: 10.3389/fdata.2022.979465

Table 1.

Recap of the suggested metrics for evaluating results of binary classifications and regression analyses.

Analysis type Always include We suggest to include
TPR, TNR, PPV, NPV, accuracy,
Binary classification MCC F1 score, Cohen's Kappa,
ROC AUC, and PR AUC
Regression analysis R2 SMAPE, MAPE, MAE, MSE, and RMSE

The formulas of the binary classification rates can be found in Chicco and Jurman (2020) and Chicco et al. (2021a,c) and the formulas of the regression analysis rates can be found in Chicco et al. (2021b).