Triple
T2494993
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Dera Ismail Khan |
E52133
|
entity |
| Predicate | administrativeFunction |
P32450
|
FINISHED |
| Object | regional administrative center |
—
|
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: regional administrative center | Statement: [Dera Ismail Khan, administrativeFunction, regional administrative center]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: administrativeFunction Context triple: [Dera Ismail Khan, administrativeFunction, regional administrative center]
-
A.
administrativeFeature
chosen
Indicates that one entity serves as an administrative or governance-related feature, function, or attribute associated with another entity.
-
B.
organizationalFunction
Indicates the role, purpose, or primary activity that an organization performs within a broader system or context.
-
C.
governmentalFunction
Indicates that an entity performs, embodies, or is responsible for an official role, duty, or activity of a government.
-
D.
administersFrom
Indicates that an action, service, or process is carried out or managed starting from a particular source, location, or authority.
-
E.
administeredAs
Indicates that one entity is given or applied to another entity as a treatment, dose, or intervention.
- 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_69ab4955111c8190835bf619adec21ff |
completed | March 6, 2026, 9:38 p.m. |
| NER | Named-entity recognition | batch_69abd19541048190b9e39db119c20fe8 |
completed | March 7, 2026, 7:19 a.m. |
| PD | Predicate disambiguation | batch_69abd0b980b481908d4932bcea4a6167 |
completed | March 7, 2026, 7:16 a.m. |
Created at: March 6, 2026, 9:45 p.m.