Table 6. Evaluation of Bi-LSTM-CNN without GloVe embedding.
| Model A without GloVe | Model B without GloVe | |||||
|---|---|---|---|---|---|---|
| – | Precision | Recall | F1-score | Precision | Recall | F1-score |
| Negative | 0.98 | 0.97 | 0.97 | 0.90 | 0.89 | 0.90 |
| Neutral | 0.92 | 0.90 | 0.91 | 0.88 | 0.90 | 0.89 |
| Positive | 0.99 | 0.99 | 0.87 | 0.88 | 0.90 | 0.89 |
| Accuracy | 0.98 | 0.91 | ||||
| Macro avg | 0.96 | 0.96 | 0.96 | 0.86 | 0.80 | 0.82 |
| Weighted avg | 0.98 | 0.98 | 0.98 | 0.87 | 0.87 | 0.87 |
| Accuracy | 0.971581 | 0.874666 | ||||
| Cohen-score | 0.941360 | 0.958807 | ||||