Triple
T8739066
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | CRPF |
E207456
|
entity |
| Predicate | RAFRole |
P84579
|
FINISHED |
| Object | handling communal riots and crowd control |
—
|
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: handling communal riots and crowd control | Statement: [CRPF, RAFRole, handling communal riots and crowd control]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: RAFRole Context triple: [CRPF, RAFRole, handling communal riots and crowd control]
-
A.
FDARole
Indicates a role or function that an entity holds within the context of FDA-related responsibilities, authority, or activities.
-
B.
unitRole
Indicates the functional role or purpose that a unit serves within a larger system or context.
-
C.
typeOfRole
Indicates that one entity specifies the kind or category of role that another entity holds or performs.
-
D.
referredAircraftRole
Indicates that one aircraft is mentioned or designated in relation to another aircraft’s role, function, or status within a specific context or event.
-
E.
roleInAH64
Indicates that an entity has a specific functional role or position in relation to the AH-64 attack helicopter.
- 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_69ca835a03a081909d4d4cd01a18c9fb |
completed | March 30, 2026, 2:06 p.m. |
| NER | Named-entity recognition | batch_69cc5d470c8c81909ead395ef704c6ba |
completed | March 31, 2026, 11:48 p.m. |
| PD | Predicate disambiguation | batch_69cc457322b481908712a9630a17b954 |
completed | March 31, 2026, 10:06 p.m. |
| PDg | Predicate description generation | batch_69cc572d99bc819097f36b140c2ee1ce |
completed | March 31, 2026, 11:22 p.m. |
Created at: March 30, 2026, 6:38 p.m.