| #     | Method                 | Algorithm | Test 1                                                  | Test 2                                                  | Test 3                                                  |
|-------|------------------------|-----------|---------------------------------------------------------|---------------------------------------------------------|---------------------------------------------------------|
| 1.a   | Deep learning (2 methods) | LSTM      | Precision: 0.4087<br>Recall: 0.4031<br>F1-Score: 0.3565<br>Accuracy: 0.3859 | Precision: 0.4984<br>Recall: 0.4800<br>F1-Score: 0.4695<br>Accuracy: 0.5533 | Precision: 0.6785<br>Recall: 0.6755<br>F1-Score: 0.6735<br>Accuracy: 0.6759 |
| 1.b   |                        | CNN model | Precision: 0.3971<br>Recall: 0.3785<br>F1-Score: 0.3727<br>Accuracy: 0.4502 | Precision: 0.5316<br>Recall: 0.5593<br>F1-Score: 0.5154<br>Accuracy: 0.5587 | Precision: 0.7699<br>Recall: 0.7596<br>F1-Score: 0.7614<br>Accuracy: 0.7591 |