Skip to main content

Table 8 The distribution of entities with various length in the correct results

From: A deep learning model incorporating part of speech and self-matching attention for named entity recognition of Chinese electronic medical records

 

1–5

6–10

10–15

> 15

SM-LSTM-CRF(char)

6886

667

52

4

SM-LSTM-CRF(char + POS)

7440

485

21

3

SM-LSTM-CRF (char + Reduced POS tagging)

7335

673

59

3

Standard set

9014

1002

104

12