Research Article
A Multigranularity Text Driven Named Entity Recognition CGAN Model for Traditional Chinese Medicine Literatures
Table 4
Experimental results under different models on the Miraculous pivot and syndrome in TCM datasets.
| Model | P (%) | R (%) | F1 (%) | P (%) | R (%) | F1 (%) | Miraculous pivot | Syndrome in TCM |
| BiLSTM-CRF | 68.76 | 66.95 | 67.85 | 72.38 | 70.95 | 71.65 | BERT-BiLSTM-CRF | 76.65 | 80.72 | 78.64 | 72.33 | 74.30 | 73.30 | Roberta-c | 76.98 | 80.47 | 78.68 | 74.62 | 75.39 | 75.00 | MT-CGAN | 77.73 | 79.69 | 78.69 | 78.45 | 76.28 | 77.35 |
|
|