Triple
T10310193
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | New York City Water Tunnel No. 3 |
E241866
|
entity |
| Predicate | stage1ServiceStart |
P72290
|
FINISHED |
| Object | 1998 |
—
|
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: 1998 | Statement: [New York City Water Tunnel No. 3, stage1ServiceStart, 1998]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: stage1ServiceStart Context triple: [New York City Water Tunnel No. 3, stage1ServiceStart, 1998]
-
A.
initialStage
Indicates that one entity represents the first or starting phase in a sequence, process, or development of another entity.
-
B.
servesStage
Indicates that an entity functions as a stage or phase within a larger process, sequence, or workflow.
-
C.
firstStageName
Indicates that one entity is the initial or earliest stage name associated with another entity in a sequence or lifecycle.
-
D.
firstStageEngine
Indicates that an engine functions as the primary propulsion unit used in the first stage of a multi-stage launch vehicle or rocket.
-
E.
implementationStart
chosen
Indicates the point in time or event at which the implementation of something (such as a plan, system, or project) begins.
- 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_69d381ac38808190a8ca7457c85b625b |
completed | April 6, 2026, 9:49 a.m. |
| NER | Named-entity recognition | batch_69d4d7ccb7ec8190a538cf279e48116e |
completed | April 7, 2026, 10:09 a.m. |
| PD | Predicate disambiguation | batch_69d4d1f4f354819080b4ed4bc61bdff6 |
completed | April 7, 2026, 9:44 a.m. |
Created at: April 6, 2026, 11:47 a.m.