How to best handle synonyms, acronyms, abbreviations when doing NER with spacy? #10753
AdrianKrebs
started this conversation in
Help: Best practices
Replies: 1 comment
-
Depends on your text and what kinds of problems you want to solve - there's no universal approach to this. Do you have known abbreviations that match specific entities? Then you can use pattern IDs to give an ID to each occurrence. Do you have abbreviations that could map to multiple entities, like ABC for "American Broadcasting Corporation" or "Alcoholic Beverages Commission"? Then you'll want the Entity Linker to disambiguate abbreviations. Do you not have abbreviations yet, but need to find them in text? Then maybe you can use this paper's method to build a list, or maybe you need a coreference model. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
What's the best way for handling synonyms, acronyms, and abbreviations of an entity when doing NER? I assume that I can have some kind of lookup table that would then map to the same entity. Let's say for "The North Face" I would want to match all "TNF" as well.
Beta Was this translation helpful? Give feedback.
All reactions