Table 5 Metrics for the different machine learning models
From: Utilizing deep learning and graph mining to identify drug use on Twitter data
Models | Accuracy (%) | Precision | Recall | F1 score | AUC |
|---|---|---|---|---|---|
CNN model-A | 82.31 | 0.893 | 0.784 | 0.835 | 0.91 |
CNN model-B | 76.35 | 0.597 | 0.906 | 0.719 | 0.90 |
BERT model-A | 79.27 | 0.850 | 0.713 | 0.775 | 0.79 |
BERT model-B | 64.25 | 0.871 | 0.338 | 0.669 | 0.64 |
Decision tree | 63.40 | 0.925 | 0.584 | 0.716 | 0.68 |
SVM | 59.33 | 0.220 | 0.943 | 0.356 | 0.87 |
XGBoost | 54.90 | 0.146 | 0.847 | 0.246 | 0.71 |
Logistic-1 | 57.44 | 0.873 | 0.546 | 0.672 | 0.58 |
Logistic-2 | 54.56 | 0.954 | 0.525 | 0.677 | 0.58 |