From: Improving chemical disease relation extraction with rich features and weakly labeled data
Feature | Type | |
---|---|---|
1 | # of chemical mention | Numeric |
2 | # of disease mention | Numeric |
3 | Is chemical in title | Boolean |
4 | Is disease in title | Boolean |
5 | Is chemical in the 1st sentence of the abstract | Boolean |
6 | Is disease in the 1st sentence of the abstract | Boolean |
7 | Is chemical in the last sentence of the abstract | Boolean |
8 | Is disease in the last sentence of the abstract | Boolean |
9 | Are both of chemical and disease in the same sentence | Boolean |
10 | Is disease-chemical relation curated by CTD in the past | Boolean |
11 | Do both disease and chemical exist in the MeSH indexing in the past? | Boolean |
12 | Is any keyword around the disease, such as therapy, complicating, affect, etc. | Boolean |
13 | Is any keyword around the chemical, such as 3.0 mEg/L, mg, etc. | Boolean |
14 | Is “increase” or “decrease” around chemical | Boolean |
15 | Is “increase” or “decrease” around disease | Boolean |
16 | Is “p value” around chemical | Boolean |
17 | Is “p-value” around disease | Boolean |
18 | Is “men”, “women”, or “patient” around chemical | Boolean |
19 | Is “men”, “women”, or “patient” around disease | Boolean |