From: Improving rare disease classification using imperfect knowledge graph
HaoDaiFu | ChinaRe | |
---|---|---|
# of documents | 51,374 | 86,663 |
# of classes (diseases) | 805 | 44 |
Vocabulary size | 59,879 | 41,087 |
Average # of words/doc | 26.7 | 29.7 |
Average # of knowledge terms/doc | 10.8 | 4.0 |