[Detection] Do not allow hyphenized acronyms

bst-mug / acres

Acronym expansion module based on word embeddings and filtering rules

Apache License 2.0

1 stars 2 forks source link

Closed michelole closed 4 years ago

michelole commented 5 years ago

So far is_acronym allows hyphenized acronyms (such as "ALA-DES") and we have to manually exclude them in our detection standard as "acro/nacr".

Instead, automate such rule in the cited method.

michelole commented 4 years ago

May not be necessary, since our cleaning routine already removes hyphens. So maybe only cleaned acronyms are valid?