Table 4.
Accuracy, precision, recall, and F1-score for four tasks using the condition and quality classification.
| Validation Set (Within-Participant) | Test Set (Leave-One-Person-Out) | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Task # | Accuracy | Precision | Recall | F1-Score | Accuracy | Precision | Recall | F1-Score | |
| Using condition classification | Task 1 | 94.8% | 93.4% | 94.3% | 93.9% | 84.9% | 80.0% | 82.3% | 81.0% |
| Task 2 | 93.6% | 91.0% | 89.3% | 90.2% | 81.1% | 67.4% | 68.7% | 67.9% | |
| Task 3 | 87.4% | 85.0% | 86.7% | 85.8% | 58.4% | 53.3% | 57.0% | 54.7% | |
| Task 4 | 91.1% | 86.4% | 87.7% | 87.0% | 73.2% | 62.0% | 63.3% | 62.5% | |
| Using quality classification | Task 1 | 93.9% | 88.6% | 92.0% | 90.1% | 80.8% | 71.0% | 76.0% | 72.7% |
| Task 2 | 88.6% | 83.8% | 80.0% | 81.7% | 76.5% | 62.0% | 62.0% | 62.0% | |
| Task 3 | 85.6% | 79.5% | 82.3% | 80.8% | 63.0% | 55.4% | 57.0% | 55.8% | |
| Task 4 | 87.2% | 81.4% | 81.0% | 81.1% | 70.9% | 60.1% | 59.7% | 59.5% | |