Triple
T22021557
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Arrondissement of Dunkirk |
E543857
|
entity |
| Predicate | contains |
P35
|
FINISHED |
| Object | Gravelines |
—
|
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: Gravelines | Statement: [Arrondissement of Dunkirk, contains, Gravelines]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Gravelines Context triple: [Arrondissement of Dunkirk, contains, Gravelines]
-
A.
Gravelines
chosen
Gravelines is a coastal commune in northern France known for its historic fortifications and strategic position along the English Channel.
-
B.
Boulogne
Boulogne is a French football club known for being one of the early professional teams in N’Golo Kanté’s career.
-
C.
Perlée
Perlée is a Van Cleef & Arpels jewelry collection distinguished by its delicate gold bead motifs and refined, playful elegance.
-
D.
Aire-sur-la-Lys
Aire-sur-la-Lys is a historic town in northern France’s Pas-de-Calais department, known for its medieval architecture and strategic location near the Lys River.
-
E.
Boulogne-sur-Mer
Boulogne-sur-Mer is a coastal city and major fishing port in northern France, located on the English Channel in the Pas-de-Calais department.
- F. None of above.
- G. Unsure - the case is ambiguous/there is not enough information to decide.
Provenance (2 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_69e11e2e8ea4819084210fe06d3a1b8d |
completed | April 16, 2026, 5:36 p.m. |
| NER | Named-entity recognition | batch_69f127c7b3308190ac056bef6f82722e |
completed | April 28, 2026, 9:33 p.m. |
Created at: April 16, 2026, 8:23 p.m.