Skip to main content
. 2020 Aug 6;8(8):e16948. doi: 10.2196/16948

Table 6.

Illustration of a Best clinical winner with max (n4)=8 for CBOW and disease target x = epilepsy, which is not an NER winner.

Ranka Candidate y 3CosAdd UMLS CUI for concept YTx mapped to candidate yTx Evidence-based categories for concept YTx correlated with concept X
1 lamotrigine 0.385201 C0064636 Tx with therapeutic effect
2 carbamazepine 0.345227 C0006949 Tx with unwanted or adverse effects (ie, nontherapeutic)
3 low_propensity 0.324285 b
4 clonazepam 0.310706 C0009011 Tx with uncertain therapeutic effect
5 topiramate 0.308402 C0076829 Tx with therapeutic effect
6 lithium_valproate 0.308223 C0023870|C0080356 Tx with therapeutic effect|Tx with unwanted or adverse effects (ie, nontherapeutic)
7 clobazam 0.306901 C0055891 Tx with therapeutic effect
8 sodium_valproate 0.300513 C0037567 Tx with therapeutic effect
9 lorazepam 0.29562 C0024002 Tx with therapeutic effect
10 lithium 0.294804 C0023870 Tx with therapeutic effect|Tx with unwanted or adverse effects (ie nontherapeutic)
11 gabapentin_pregabalin_topiramate 0.291698 C0657912|C0076829|C0060926 Tx with therapeutic effect
12 antiepileptic_drugs_other_than 0.290046 C0003299 Tx with therapeutic effect

aThe search query −x +z1 +z2 is listed in Table 4, which is −epilepsy +valproate +antiepileptic_drug. The character “|” appears when there is more than 1 CUI or evidence-based category.

bThe candidate y = “low_propensity” does not belong to the semantic field Tx, and so, it has no UMLS CUI assigned.