Triple
T7440302
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Savaiʻi |
E171731
|
entity |
| Predicate | countryCapitalConnection |
P76415
|
FINISHED |
| Object | linked by ferry to Apia on Upolu |
—
|
LITERAL 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: linked by ferry to Apia on Upolu | Statement: [Savaiʻi, countryCapitalConnection, linked by ferry to Apia on Upolu]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: countryCapitalConnection Context triple: [Savaiʻi, countryCapitalConnection, linked by ferry to Apia on Upolu]
-
A.
countryCapitalRelation
Indicates that one entity is the capital city of the country represented by the other entity.
-
B.
countryCapitalAssociation
Indicates the relationship in which a specific city serves as the capital of a particular country.
-
C.
connectsCapitalOf
Indicates a relationship where one entity serves as the capital city of another entity, linking the capital to the political or administrative unit it represents.
-
D.
countryCapitalContext
Indicates that one entity serves as the capital city of the specified country in a given contextual or temporal setting.
-
E.
connectsCapitalCity
Indicates a relationship where an entity is linked or associated specifically with a capital city.
- F. None of above. chosen
Provenance (4 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_69c68a64228c8190affaec2a8127ce7b |
completed | March 27, 2026, 1:47 p.m. |
| NER | Named-entity recognition | batch_69c6f34d84008190936af2b3670ef210 |
completed | March 27, 2026, 9:14 p.m. |
| PD | Predicate disambiguation | batch_69c6f038582c8190bac77c9b5a34b862 |
completed | March 27, 2026, 9:01 p.m. |
| PDg | Predicate description generation | batch_69c6f0be2b1c8190bea06100a7caef2b |
completed | March 27, 2026, 9:03 p.m. |
Created at: March 27, 2026, 3:13 p.m.