Triple
T7532585
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Canvey Island |
E178061
|
entity |
| Predicate | hasSeaWall |
P34503
|
FINISHED |
| Object | Canvey Island sea defences |
—
|
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: Canvey Island sea defences | Statement: [Canvey Island, hasSeaWall, Canvey Island sea defences]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: hasSeaWall Context triple: [Canvey Island, hasSeaWall, Canvey Island sea defences]
-
A.
hasSeaAccess
Indicates that an entity has direct access to the sea, typically via a coastline, port, or navigable waterway connected to the sea.
-
B.
hasBreakwater
chosen
Indicates that a structure or location is equipped with or associated with a breakwater that provides protection from waves or erosion.
-
C.
hasHighwayAlongShore
Indicates that a highway runs adjacent to or closely follows the shoreline of a body of water.
-
D.
touchesSea
Indicates that one entity is in direct physical contact with the sea or coastline of another entity.
-
E.
hasSeaCondition
Indicates that an entity is associated with or characterized by a particular state or condition of the sea.
- F. None of above.
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_69c69f2acdbc8190b5a8320168c1d0ba |
completed | March 27, 2026, 3:15 p.m. |
| NER | Named-entity recognition | batch_69c6f84753fc81908bee2013004ef5fb |
completed | March 27, 2026, 9:36 p.m. |
| PD | Predicate disambiguation | batch_69c6f4d8eedc81908c1ae421e0e63798 |
completed | March 27, 2026, 9:21 p.m. |
Created at: March 27, 2026, 3:47 p.m.