Triple
T26535211
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Tuas Checkpoint |
E671225
|
entity |
| Predicate | sisterCheckpoint |
P61733
|
FINISHED |
| Object | Woodlands Checkpoint |
—
|
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: Woodlands Checkpoint | Statement: [Tuas Checkpoint, sisterCheckpoint, Woodlands Checkpoint]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: sisterCheckpoint Context triple: [Tuas Checkpoint, sisterCheckpoint, Woodlands Checkpoint]
-
A.
sisterChannel
Indicates that one channel is a sibling or counterpart to another channel, typically under the same ownership or network.
-
B.
sisterTerminal
chosen
Indicates that two terminals share a sibling-like relationship, typically meaning they are distinct but related endpoints within the same larger system or structure.
-
C.
sisterComplex
Indicates a strong, often excessive or romanticized emotional fixation or attraction that someone has toward their sister.
-
D.
sisterPillar
Indicates a relationship where one pillar is considered the sister (a closely related or counterpart structure) to another pillar.
-
E.
sisterBase
Indicates that one entity is the sister of another, sharing at least one parent and being female relative to the other entity.
- 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_69eeb3206e748190b90c85cc81f38c91 |
completed | April 27, 2026, 12:51 a.m. |
| NER | Named-entity recognition | batch_69f613fb435c8190b586d8a73880f673 |
completed | May 2, 2026, 3:10 p.m. |
| PD | Predicate disambiguation | batch_69f602d7b1b0819095ddd3b5169f8ce2 |
completed | May 2, 2026, 1:57 p.m. |
Created at: April 27, 2026, 1:38 a.m.