Triple
T10883658
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Alaska Time |
E256985
|
entity |
| Predicate | offsetFromUTCDaylightTime |
P25647
|
FINISHED |
| Object | UTC−08:00 |
—
|
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: UTC−08:00 | Statement: [Alaska Time, offsetFromUTCDaylightTime, UTC−08:00]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: offsetFromUTCDaylightTime Context triple: [Alaska Time, offsetFromUTCDaylightTime, UTC−08:00]
-
A.
offsetFromUTC
chosen
Indicates the time difference between a given time value and Coordinated Universal Time (UTC), typically expressed as an offset in hours and/or minutes.
-
B.
offsetFromUTCInMinutes
Indicates the number of minutes by which a given time value differs from Coordinated Universal Time (UTC), positive or negative.
-
C.
offsetFromUTCInHours
Indicates the number of hours a time value is offset from Coordinated Universal Time (UTC), positive for time zones ahead and negative for those behind.
-
D.
offsetInSecondsFromUTC
Indicates the time difference, measured in whole seconds, between a given time reference and Coordinated Universal Time (UTC).
-
E.
offsetFromJapanStandardTimeHours
Indicates the number of hours by which a time value is ahead of or behind Japan Standard Time.
- 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_69d6aa848804819081b2713ca0bedf06 |
completed | April 8, 2026, 7:20 p.m. |
| NER | Named-entity recognition | batch_69d751dc3518819090e03d81ea5c8aa9 |
completed | April 9, 2026, 7:14 a.m. |
| PD | Predicate disambiguation | batch_69d70d360c388190a3d829fe8862434f |
completed | April 9, 2026, 2:21 a.m. |
Created at: April 8, 2026, 9:21 p.m.