Table 6 The performance of JLAN and the baseline models on the MIMIC-III-full test set
From: JLAN: medical code prediction via joint learning attention networks and denoising mechanism
Model | AUC | Â | F1 | Â | P@15 | P@8 |
|---|---|---|---|---|---|---|
| Â | Macro | Micro | Macro | Micro | Â | Â |
LR | 56.1 | 93.7 | 1.1 | 27.2 | – | 54.2 |
CNN | 80.6 | 96.9 | 4.2 | 41.9 | – | 58.1 |
BiGRU | 82.2 | 97.1 | 3.8 | 41.7 | Â | 58.5 |
CAML | 89.5 | 98.6 | 8.8 | 53.9 | – | 70.9 |
DR-CAML | 89.7 | 98.5 | 8.6 | 52.9 | – | 69.0 |
MSATT-KG | 91.0 | 99.2 | 9.0 | 55.3 | – | 72.8 |
MultiResCNN | 91.0 | 98.6 | 8.5 | 55.2 | 58.4 | 73.4 |
JLAN | 91.8 | 98.8 | 9.7 | 56.7 | 57.9 | 74.1 |