Triple
T28262183
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | deserts the army for Carmen |
E712610
|
entity |
| Predicate | hasCharacterAgent |
P19389
|
FINISHED |
| Object | Don José |
—
|
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: Don José | Statement: [deserts the army for Carmen, hasCharacterAgent, Don José]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: hasCharacterAgent Context triple: [deserts the army for Carmen, hasCharacterAgent, Don José]
-
A.
hasAgent
chosen
Indicates that an action or event is carried out or initiated by a particular agent.
-
B.
hasHumanCharacters
Indicates that the subject includes or features characters that are human beings.
-
C.
hasCharacters
Indicates that an entity (such as a work or story) includes or features certain characters as part of its content.
-
D.
hasHumanCharacterRole
Indicates that an entity is assigned a role or function specifically associated with a human character within a context such as a story, performance, or representation.
-
E.
hasCharacterControl
Indicates that one entity has the ability or authority to direct, manipulate, or govern the behavior or state of 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_69efb5207eb08190827e4c34048030b1 |
completed | April 27, 2026, 7:12 p.m. |
| NER | Named-entity recognition | batch_69f74062b9388190b30546cf700a825c |
completed | May 3, 2026, 12:32 p.m. |
| PD | Predicate disambiguation | batch_69f73c802b848190b61a416b7488bd96 |
completed | May 3, 2026, 12:16 p.m. |
Created at: April 27, 2026, 11:12 p.m.