Triple
T32148078
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Saville Dam |
E821074
|
entity |
| Predicate | nearCommunityRelocatedForReservoir |
P4647
|
FINISHED |
| Object | Barkhamsted Hollow |
—
|
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: Barkhamsted Hollow | Statement: [Saville Dam, nearCommunityRelocatedForReservoir, Barkhamsted Hollow]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: nearCommunityRelocatedForReservoir Context triple: [Saville Dam, nearCommunityRelocatedForReservoir, Barkhamsted Hollow]
-
A.
residenceAfterFlood
Indicates that an entity’s place of residence is the one occupied after a flood event has occurred.
-
B.
relatedReservoir
Indicates that one reservoir has a specified relationship or association with another reservoir, such as physical connection, shared resources, or functional linkage.
-
C.
residesNear
Indicates that one entity lives or is located in close physical proximity to another entity.
-
D.
reservoirLocation
Indicates that a reservoir is geographically located at or within a specified place or region.
-
E.
hasNearbyCommunity
chosen
Indicates that one entity has another community located close to it in geographic or spatial terms.
- 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_69f3490520d081909b2f1271dab75faa |
completed | April 30, 2026, 12:20 p.m. |
| NER | Named-entity recognition | batch_69f6b9e567908190abe980a80db9a7af |
completed | May 3, 2026, 2:58 a.m. |
| PD | Predicate disambiguation | batch_69f6b3a970b0819090c6473844ffa8e3 |
completed | May 3, 2026, 2:32 a.m. |
Created at: May 1, 2026, 12:31 a.m.