Skip to main content

Table 5 Evaluation Metrics for Calibrating Probabilities for Each Model

From: Classification of imbalanced data using machine learning algorithms to predict the risk of renal graft failures in Ethiopia

Algorithms

AUC-ROC

Precision

Recall

F

Brier Score

LR

0.7272727

0.4000000

0.3333333

0.3636364

0.05823289↑

NB

0.7857143↑

0.3333333↑

0.5000000↑

0.4000000↑

0.05954727↑

ANN

0.7175325

0.7500000↑

0.5000000

0.6000000↑

0.04688012↑

RF

0.8365801↑

0.6666667↑

0.3333333

0.4444444↑

0.04498363↑

TBAG

0.8376623

0.3076923

0.6666667

0.4210526

0.05800087↑

SGB

0.7889610↑

0.5000000

0.5000000

0.5000000

0.04895868↑

  1. Bold: column’s best estimates, ↑: shows improvement from previous approaches