Table 6 Model improvement results under the CCKS2019 dataset.
From: Chinese medical named entity recognition integrating adversarial training and feature enhancement
Models | Precision | Recall | F1 Score |
|---|---|---|---|
BiLSTM-CRFbase | 82.69% | 82.44% | 82.55% |
BiLSTM-IDCNN-CRF | 85.64% | 84.69% | 85.16% |
BERT-BiLSTM-CRF | 91.26% | 90.88% | 91.06% |
BERT-BiLSTM-IDCNN-CRF | 91.28% | 91.31% | 91.30% |
BPBIC | 91.57% | 91.97% | 91.77% |
