Fusion Deep Learning and Machine Learning for Heterogeneous Military Entity Recognition

Table 5

Experimental results of the BERT-BiLSTM-CRF model on corpus sets with different size.

	(%)	(%)	(%)
EP-ONE	75.03	72.14	73.56
EP-TWO	81.29	79.06	80.16
EP-THREE	85.73	83.37	84.53