Table 4 Performances of our CNN-LSTM-Attention model on each category under “strict” criterion
From: Entity recognition in Chinese clinical text using attention-based CNN-LSTM-CRF
Category | ICRC_CNER (%) | CCKS2017_CNER (%) | ||||
|---|---|---|---|---|---|---|
Pre. | Rec. | F1 | Pre. | Rec. | F1 | |
Disease | 82.84 | 81.67 | 82.25 | 85.06 | 77.22 | 80.95 |
Symptom | 77.06 | 76.01 | 76.53 | 94.92 | 96.28 | 95.60 |
Test | 84.19 | 89.03 | 86.55 | 93.66 | 93.48 | 93.57 |
Treatment | 77.53 | 79.58 | 78.54 | 77.63 | 79.08 | 83.98 |
Medication | 87.88 | 89.72 | 88.79 | / | / | / |
Body | / | / | / | 86.89 | 87.36 | 87.12 |