Skip to main content

Table 15 Sample tokens and their chemical segment composition.

From: Optimising chemical named entity recognition with pre-processing analytics, knowledge-rich features and heuristics

Token initially recognised as non-chemical

Chemical basic segments

Ratio

polycalcium

poly, calcium

1.0

2-methoxyestradiol

meth, oxy, estra, di, ol

0.89

palytoxin

toxin

0.56