TCMNER and PubMed: A Novel Chinese Character-Level-Based Model and a Dataset for TCM Named Entity Recognition

Table 3

The statistics of the PubMed dataset.

	Train set	Valid set	Test set
Clinical manifestation	24332	8111	8111
Syndrome	7613	2538	2538
Disease	2808	935	935
Treatment law	11186	3728	3728
Herb	11682	3627	3627
Total	56,628	18,876	18,876