From: Comparison of machine learning techniques to predict all-cause mortality using fitness data: the Henry ford exercIse testing (FIT) project
Single Normal
kernel Estimation
Supervised Discretization
Sensitivity
35.32%
40.90%
37.41%
Specificity
93.26%
92.37%
93.32%
Precision
52.34%
42.70%
52.20%
F-score
42.18%
41.78%
43.59%
RMSE
0.35
0.32
0.34
AUC
0.81
0.82