EntityTypeOf()
The rule-based NER above detects + coarsely types spans. These re-TYPE an entity against ARBITRARY, user-defined types BY MEANING (zero-shot, via the neural model when loaded; lexical fallback otherwise) -- so the type set adapts to any domain, and "Paris" the city vs the person is disambiguated by meaning. (A real token-classification NER-head GGUF, once available, plugs onto the per-token engine substrate -- neural_embed_tokens -- underneath.) EntityTypeOf(entity, types) -- the candidate type whose MEANING best matches the entity mention (DATA, a string).
Basic examples (1)
t = new stzText("Barack Obama loves the wonderful city of Paris.")
# EntityTypeOf picks the closest type
? t.EntityTypeOf("New York City", [ "city location", "food dish" ])
# a non-list of types is rejected
? t.EntityTypeOf("Paris", "not a list")
ae = t.EntitiesTypedAs([ "person", "place" ])
# EntitiesTypedAs returns [entity, type] pairs for detected entities
? len(ae) >= 1 and len(ae[1]) = 2
# EntitiesTypedAsQQ is stzListOfPairs
? classname(t.EntitiesTypedAsQQ([ "person", "place" ]))city location 1 stzlistofpairs
run on 2026-10-02 04:23 inside the library at commit 0e72e2e2c; taken from text_query_layer_narrated.ring(opens in a new tab)
The extensions of this method
Each extension, and whether it exists for this method; an extension that exists combines with the others (FindSTCS).
Every example is a test file of the library, run inside the library for this page; only those whose output kept every promise their file wrote are shown.
See also
- The class stzText
- The area's reference: NLP & linguistics
- The other methods in these examples: stzText.EntitiesTypedAs

