Table 3.
Models | AUC | Precision | Recall | F1 score (Online supplementary file 1) |
Bayesian network | 0.764±0.029 | 0.729±0.055 | 0.717±0.042 | 0.721±0.025 |
KNN | 0.838±0.018 | 0.813±0.046 | 0.636±0.062 | 0.712±0.039 |
SVM | 0.765±0.044 | 0.728±0.059 | 0.589±0.078 | 0.647±0.046 |
C&R Tree | 0.755±0.030 | 0.739±0.026 | 0.669±0.066 | 0.700±0.033 |
CHAID | 0.770±0.041 | 0.792±0.035 | 0.655±0.049 | 0.716±0.031 |
Ensemble | 0.866±0.082 | 0.824±0.043 | 0.732±0.061 | 0.773±0.032 |
P value | p<0.0001 | p<0.0001 | p<0.0001 | p<0.0001 |
The bold value indicate the performance parameters of the best algorithm.
AUC, area under the receiver operating characteristic curve.