Triple
T21242222
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Sinking of ROKS Cheonan |
E523502
|
entity |
| Predicate | hullBreakType |
P56828
|
FINISHED |
| Object | catastrophic break in two |
—
|
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: catastrophic break in two | Statement: [Sinking of ROKS Cheonan, hullBreakType, catastrophic break in two]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: hullBreakType Context triple: [Sinking of ROKS Cheonan, hullBreakType, catastrophic break in two]
-
A.
breakType
chosen
Indicates the specific kind or category of break or interruption that occurs in a process, activity, or sequence.
-
B.
brokenBy
Indicates that one entity causes the damage, destruction, or loss of functionality of another entity.
-
C.
breaksOver
Indicates that one entity moves or crashes forcefully onto the surface or boundary of another entity, typically dispersing or shattering upon contact.
-
D.
hasHullType
Indicates that an entity possesses or is characterized by a specific type or form of hull.
-
E.
breaksDownWhen
Indicates that one entity becomes nonfunctional, damaged, or fails when subjected to the presence, action, or conditions imposed by another 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_69e0b513b89c81908b27147e91368db2 |
completed | April 16, 2026, 10:08 a.m. |
| NER | Named-entity recognition | batch_69e7352453f88190b38de7b0108c683a |
completed | April 21, 2026, 8:28 a.m. |
| PD | Predicate disambiguation | batch_69e5f61239708190ab7b3c83ae848a0d |
completed | April 20, 2026, 9:46 a.m. |
Created at: April 16, 2026, 3:47 p.m.