Table 1 Performance of the top classifier (Random Forest) obtained from Auto-Weka using 10-fold cross-validation and bootstrap sampling for diabetes, hypertension, and mortality datasets.

From: FIT calculator: a multi-risk prediction framework for medical outcomes using cardiorespiratory fitness data

 

10-fold cross-validation

Bootstrap sampling

RF diabetes

RF hypertension

RF mortality

RF diabetes

RF hypertension

RF mortality

Sensitivity

0.82 ± 0.002

0.83 ± 0.003

0.92 ± 0.002

0.82 ± 0.002

0.84 ± 0.003

0.91 ± 0.002

Accuracy

0.82 ± 0.001

0.83 ± 0.002

0.92 ± 0.003

0.82 ± 0.001

0.83 ± 0.001

0.92 ± 0.003

Precision

0.82 ± 0.002

0.83 ± 0.001

0.92 ± 0.001

0.82 ± 0.001

0.82 ± 0.002

0.92 ± 0.001

F-Score

0.82 ± 0.002

0.82 ± 0.002

0.92 ± 0.002

0.83 ± 0.002

0.83 ± 0.003

0.93 ± 0.002

RMSE

0.35 ± 0.001

0.34 ± 0.002

0.25 ± 0.001

0.32 ± 0.002

0.33 ± 0.002

0.26 ± 0.002

AUC

0.89 ± 0.001

0.90 ± 0.001

0.90 ± 0.002

0.89 ± 0.001

0.89 ± 0.001

0.90 ± 0.001

Back to article page