Triple
T15795236
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Georgian highlands |
E382959
|
entity |
| Predicate | language |
P15
|
FINISHED |
| Object | Svan |
E278111
|
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: Svan | Statement: [Georgian highlands, language, Svan]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Svan Context triple: [Georgian highlands, language, Svan]
-
A.
Svan people
chosen
The Svan people are an indigenous ethnic subgroup of Georgians from the remote, mountainous Svaneti region of northwestern Georgia, known for their distinct culture, traditions, and historical tower villages.
-
B.
Svaliava
Svaliava is a small town in western Ukraine known for its scenic Carpathian surroundings and mineral springs.
-
C.
Alpaida
Alpaida was a noble Frankish woman of the early 8th century best known as the consort of Pepin of Herstal and the mother of Charles Martel.
-
D.
Svane
Svane is a Danish surname most notably associated with Mikkel Svane, the co-founder and former CEO of the customer service software company Zendesk.
-
E.
Valka
Valka is a compassionate and fiercely independent dragon rider who serves as Hiccup’s long-lost mother and a key protector of dragons in the How to Train Your Dragon film series.
- 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_69d86da16e188190b89af699f1ed0bfe |
completed | April 10, 2026, 3:25 a.m. |
| NER | Named-entity recognition | batch_69e0b4dc887081909d682ae153f06d97 |
completed | April 16, 2026, 10:07 a.m. |
| NED1 | Entity disambiguation (via context triple) | batch_69ff90aea81c8190ad8bc0cdedf4b77a |
completed | May 9, 2026, 7:53 p.m. |
Created at: April 10, 2026, 4:48 a.m.