Triple
T28331027
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Nation station |
E717538
|
entity |
| Predicate | connectsArrondissements |
P193594
|
FINISHED |
| Object | 11th arrondissement of Paris |
—
|
NE NERFINISHED |
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: 11th arrondissement of Paris | Statement: [Nation station, connectsArrondissements, 11th arrondissement of Paris]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: connectsArrondissements Context triple: [Nation station, connectsArrondissements, 11th arrondissement of Paris]
-
A.
hasArrondissement
Indicates a relationship where an administrative unit or locality is associated with, or belongs to, a specific arrondissement.
-
B.
connectsCentralDistrictsWith
Indicates a relationship where something serves as a link or route joining central districts to one another.
-
C.
hasNumberOfArrondissements
Indicates the relationship that specifies how many arrondissements (administrative districts) are associated with a given entity.
-
D.
capitalOfArrondissement
Indicates that a place serves as the administrative capital (seat) of a specified arrondissement.
-
E.
hasAdjacentArrondissement
Indicates that one arrondissement is directly next to or shares a boundary with another arrondissement.
- 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_69eff6e9a57c8190a69c2c74b5d72119 |
completed | April 27, 2026, 11:53 p.m. |
| NER | Named-entity recognition | batch_69fd4d1854988190be093b103a681798 |
completed | May 8, 2026, 2:40 a.m. |
| PD | Predicate disambiguation | batch_69fd4c8d1a188190897c24527337814a |
completed | May 8, 2026, 2:38 a.m. |
| PDg | Predicate description generation | batch_69fd4d16dd20819096957c40f43cd971 |
completed | May 8, 2026, 2:40 a.m. |
Created at: April 28, 2026, 12:32 a.m.