lasigeBioTM / MER

Minimal Named-Entity Recognizer (MER)
http://labs.fc.ul.pt/mer/
56 stars 8 forks source link

Missing annotation when word precedes a comma #13

Closed LLCampos closed 7 years ago

LLCampos commented 7 years ago

When running:

bash get_entities.sh 1 A 'water potassium oxygen' HMDB_ChEMBL_ChEBI

We get:

1   A   0   5   0.378665    water   unknown 1
1   A   6   15  0.54488 potassium   unknown 1
1   A   16  22  0.441889    oxygen  unknown 1

But when we run:

bash get_entities.sh 1 A 'water, potassium, oxygen' HMDB_ChEMBL_ChEBI

We get:

1   A   0   5   0.378665    water   unknown 1
1   A   18  24  0.441889    oxygen  unknown 1

Strangely, "water" is annotated but "potassium" isn't, although both precede a comma.