Table 8 Performance evaluation on real-word error correction task
From: Improving the quality of Persian clinical text with a novel spelling correction system
Task | \(\mathbf{M}\mathbf{o}\mathbf{d}\mathbf{e}\mathbf{l}\) | Precision | Recall | F1-Score |
|---|---|---|---|---|
Real-word Error Detection: | Pre-trained Model | 90.2 | 91.1 | 90.6 |
Real-word Error Correction: | Pre-trained Model | 89.4 | 90.7 | 90.0 |
Pre-trained Model + PERTO | 90.8 | 92.2 | 91.5 |