Research Article
MAF-CNER : A Chinese Named Entity Recognition Model Based on Multifeature Adaptive Fusion
Table 3
Comparison results between the model in this article and the basic model.
| | MSRA | People’s Daily | | Model | Precision (%) | Recall (%) | F1 (%) | Precision (%) | Recall (%) | F1 (%) |
| | LSTM | 84.56 | 83.25 | 83.90 | 87.25 | 85.76 | 86.49 | | BiLSTM | 86.77 | 84.93 | 85.84 | 88.34 | 87.61 | 87.97 | | BiLSTM-CRF | 89.45 | 87.20 | 88.31 | 90.45 | 89.72 | 90.08 | | Ours | 97.21 | 96.83 | 97.01 | 97.08 | 96.47 | 96.78 |
|
|
