Table 2 Distribution of CCKS2019 data training and test sets. The CCKS2019 + dataset will be divided into training set and test set according to 8:2. Table 3 gives the distribution of various types of entities in the CCKS2019 + dataset training and test sets.

From: Chinese medical named entity recognition integrating adversarial training and feature enhancement

CCKS2019

Dsa

Chk

Ins

Sur

Med

Ana

Train

2116

222

318

765

456

1486

Test

682

91

193

140

263

447