umcu / clinlp

A Python library for performing NLP on clinical text written in Dutch
GNU General Public License v3.0
33 stars 0 forks source link

Add term exclusion list functionality to Entity Matcher #66

Open bramiozo opened 3 months ago

bramiozo commented 3 months ago

On top of the pseudo type, it would be nice to have the option to be able to add a term exclusion list. Especially if you set the "proximity" to >0 it may introduce words that change the semantics. Easy fix would be to add an exclusion list.

Complication: it could be that a term in the exclusion list is also explicitly present in the concept dictionary, in which case the exclusion term should have no effect.

A typical example I encounter is when I add a list of phrases indicative of the presence of a medical concept. I would like to add a proximity of >0 to improve the generalisation but this opens up the possibility an explicit denial term creeping in the phrase, hence I would like to add an exclusion list of denial terms like "geen".