Named Entity Recognition - Model: BERT-MRC

The BERT-MRC model is used in the field of entity recognition. The effect is better than other models when the amount of data is small. The reason is that the BERT-MRC model can add some prior knowledge through the problem to reduce the amount of data.The problem caused by too small is in the actual experiment, in the case of a relatively small amount of data, the effect of the BERT-MRC model is better than other models.The BERT-MRC model is suitable for use in scenarios where labeled data is lacking.

---

---

---

References:
Named Entity Recognition Baseline Model BERT-MRC Summary