�

List of performance metrics

Classification tasks

Metrics that built around confusion matrix:

[X] Confusion Matrix (True Positive, True Negative, False Positive - Type I Error, False Negative - Type II Error)
[X] True Positive Rate (TPR) / Hit Rate / Recall / Sensitivity / Detection Rate
[X] True Negative Rate (TNR) / Specificity / Selectivity
[X] False Positive Rate (FPR) / Fall-out / False Alarm Rate (FAR)
[X] False Negative Rate (FNR) / Miss Rate
[X] Accuracy / Correct Classification Rate (Correctly Classified Rate)
[X] Misclassification Rate
[X] Balanced Accuracy
[ ] Balanced Error Rate
[X] Positive Predicted Value (PPV) / Precision
[ ] Average Precision
[X] False Discovery Rate (FDR)
[X] False Omission Rate (FOR)
[ ] Positive Likelihood
[ ] Negative Likelihood
[X] Prevalence
[X] F1 Score
[X] Matthews Correlation Coefficient (MCC)
[ ] Discriminant Power
[X] Informedness (Bookmaker Informedness - BM) / Youden Index (Youden’s J Statistic)
[X] Markedness (MK)
[X] AUROC (Area Under ROC)
[X] AR (Accuracy Ratio) / Gini Coefficient
[X] KS Statistic
[ ] AUPR (Area Under Precision-Recall Curve)
[ ] Bayesian Error Rate
[X] Fowlkes-Mallows Index / G-Score
[X] Cohen Kappa
[ ] Hamming Loss
[ ] Hinge Loss
[ ] Jaccard Score

Proper scoring rule: