Triple
T8879966
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Abstract Wikipedia |
E211383
|
entity |
| Predicate | hasAbbreviation |
P43
|
FINISHED |
| Object | AW |
E188331
|
NE FINISHED |
How this triple was built (2 steps)
Every LLM step that produced this triple, in pipeline order — named-entity classification, the disambiguation choices (the exact options shown, with the pick highlighted), and the generated description. The batch + timestamp of each is in the Provenance table below.
NER
Named-entity recognition
gpt-5-mini
Instruction
Given a phrase, classify it is english named entity (e.g., persons, organizations, works of art) in Latin script, or not (e.g., literals, dates, URLs, verbose phrases). For disambiguation, the statement where the phrase occurs as object is also given. Please return a JSON object with `phrase` (string, the phrase being analyzed) and `is_ne` (boolean, indicating whether the phrase is a Named Entity).
Input
Phrase: AW | Statement: [Abstract Wikipedia, hasAbbreviation, AW]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: AW Context triple: [Abstract Wikipedia, hasAbbreviation, AW]
-
A.
AW
chosen
AW is the two-letter ISO 3166-1 alpha-2 country code assigned to Aruba.
-
B.
AW
AW is the standard abbreviation for Atlantic Water, the relatively warm, salty water mass originating in the North Atlantic that influences ocean circulation and Arctic climate.
-
C.
AW
AW is a German vehicle registration code associated with the district that includes the town of Remagen in the state of Rhineland-Palatinate.
-
D.
Wa
Wa is a town in northwestern Ghana that serves as an administrative, commercial, and cultural hub for the surrounding region.
-
E.
Wa
Wa is an ancient Chinese exonym historically used to refer to the people and early polities of the Japanese archipelago.
- F. None of above.
- G. Unsure - the case is ambiguous/there is not enough information to decide.
Provenance (3 batches)
The batch behind each pipeline step, in order, with when it ran. Timestamps are batch-level — stages were processed in waves, so the object chain (NER → NED1 → NEDg → NED2) reads in order, but predicate / elicitation batches can sit in a different wave.
| Step | Stage | Batch ID | Status | When |
|---|---|---|---|---|
| creating | Elicitation | batch_69ca838f9e20819096ab1f236a70381a |
completed | March 30, 2026, 2:07 p.m. |
| NER | Named-entity recognition | batch_69cc61677c9c8190aa09dc2a05d4cf95 |
completed | April 1, 2026, 12:05 a.m. |
| NED1 | Entity disambiguation (via context triple) | batch_69cfabc1992481909e8a4216086d5111 |
completed | April 3, 2026, noon |
Created at: March 30, 2026, 6:52 p.m.