Table 3 The performance of traditional machine learning models on the independent test set
From: Identification of plant vacuole proteins by using graph neural network and contact maps
Model | Acc (%) | F1-score | Sp (%) | Sn (%) | MCC | ROC-AUC | PR-AUC |
|---|---|---|---|---|---|---|---|
GaussianNB | 59.46 | 0.6386 | 47.30 | 71.62 | 0.1950 | 0.6291 | 0.7190 |
LR | 63.51 | 0.6582 | 56.76 | 70.27 | 0.2728 | 0.6888 | 0.6570 |
SVM | 66.89 | 0.6839 | 62.16 | 71.62 | 0.3394 | 0.7014 | 0.6796 |
RF | 66.22 | 0.6667 | 64.86 | 67.57 | 0.3244 | 0.7062 | 0.7001 |
LightGBM | 66.89 | 0.6573 | 70.27 | 63.51 | 0.3386 | 0.7144 | 0.7074 |
GBDT | 64.18 | 0.6345 | 66.22 | 62.16 | 0.2840 | 0.6843 | 0.6763 |
MLP | 62.16 | 0.6164 | 63.51 | 60.81 | 0.2433 | 0.6770 | 0.6639 |
KNN | 60.14 | 0.6144 | 56.75 | 63.51 | 0.2032 | 0.6462 | 0.6600 |
XGBoost | 63.51 | 0.5846 | 75.67 | 51.35 | 0.2786 | 0.6707 | 0.6681 |