Method | Class | Precision | Recall | F1-Score | Accuracy |
Decision Tree | Negative | 0.88 | 0.95 | 0.91 | 0.90 |
Positive | 0.94 | 0.85 | 0.89 | ||
SVC | Negative | 0.77 | 0.49 | 0.60 | 0.65 |
Positive | 0.60 | 0.83 | 0.69 | ||
Random Forest | Negative | 0.86 | 0.94 | 0.90 | 0.89 |
Positive | 0.92 | 0.82 | 0.87 | ||
Multinomial Naïve Bayes | Negative | 0.62 | 0.03 | 0.05 | 0.48 |
Positive | 0.47 | 0.98 | 0.64 | ||
LSTM (LR = 0.001) | Negative | 0.97 | 0.95 | 0.96 | 0.9599 |
Positive | 0.95 | 0.97 | 0.96 | ||
BiLSTM (LR = 0.001) | Negative | 0.96 | 0.95 | 0.96 | 0.9573 |
Positive | 0.95 | 0.96 | 0.96 | ||
GRU (LR = 0.001) | Negative | 0.97 | 0.96 | 0.97 | 0.9563 |
Positive | 0.96 | 0.97 | 0.96 | ||
BiGRU (LR = 0.001) | Negative | 0.96 | 0.95 | 0.96 | 0.9553 |
Positive | 0.95 | 0.96 | 0.96 | ||
CNN-LSTM (LR = 0.001) | Negative | 0.97 | 0.96 | 0.97 | 0.9679 |
Positive | 0.96 | 0.97 | 0.97 | ||
Attention-LSTM (LR = 0.01) | Negative | 0.97 | 0.97 | 0.97 | 0.9706 |
Positive | 0.97 | 0.97 | 0.97 |