I have encountered cases where I have brand names of entities that contain characters like ®, ™ and ©. These names that are not groundable, however when stripped of these special characters, they are groundable. One example is Advil™ vs Advil, where the former is not groundable while the latter grounds to ibuprofen CHEBI:5855.
Is it possible to add a step in the normalization that removes characters like ®, ™ and ©? If so I can submit a PR for that.
I think the issue is universal enough that it could motivate this to be added to gilda rather than being handled by the caller of the grounding function.
I have encountered cases where I have brand names of entities that contain characters like ®, ™ and ©. These names that are not groundable, however when stripped of these special characters, they are groundable. One example is
Advil™vsAdvil, where the former is not groundable while the latter grounds to ibuprofenCHEBI:5855.Is it possible to add a step in the normalization that removes characters like ®, ™ and ©? If so I can submit a PR for that.
I think the issue is universal enough that it could motivate this to be added to gilda rather than being handled by the caller of the grounding function.