Triple
T4208608
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Air France Flight 447 |
E93840
|
entity |
| Predicate | cockpitVoiceRecorderRecoveryDate |
P54756
|
FINISHED |
| Object | 2011-05 |
—
|
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: 2011-05 | Statement: [Air France Flight 447, cockpitVoiceRecorderRecoveryDate, 2011-05]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: cockpitVoiceRecorderRecoveryDate Context triple: [Air France Flight 447, cockpitVoiceRecorderRecoveryDate, 2011-05]
-
A.
operatorDuringRecordFlight
Indicates that an operator is responsible for or engaged in operating a system or vehicle during the time a particular flight is being recorded.
-
B.
cockpitCommonalityWith
Indicates that two aircraft or vehicle models share a similar or standardized cockpit layout, controls, or interface design.
-
C.
cockpitType
Indicates the specific configuration or style of cockpit associated with an entity (e.g., vehicle or aircraft).
-
D.
rocketCoreRecovery
Indicates the process or outcome of retrieving and securing the main rocket core after its launch or mission.
-
E.
launchAndRecovery
Indicates a relationship where one entity is responsible for both deploying (launching) another entity into operation and subsequently retrieving (recovering) it afterward.
- F. None of above. chosen
Provenance (4 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_69b3451743608190808f41d17ccf2650 |
completed | March 12, 2026, 10:58 p.m. |
| NER | Named-entity recognition | batch_69b34e098da881909a0cc339cc186627 |
completed | March 12, 2026, 11:36 p.m. |
| PD | Predicate disambiguation | batch_69b347efd9b08190bb50f82e4e7fe06d |
completed | March 12, 2026, 11:10 p.m. |
| PDg | Predicate description generation | batch_69b34e04ef1c81908bb34ae1cbfab1e6 |
completed | March 12, 2026, 11:36 p.m. |
Created at: March 12, 2026, 11:03 p.m.