Triple
T6282773
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Norfolk four-course rotation |
E140818
|
entity |
| Predicate | componentPractice |
P69868
|
FINISHED |
| Object | growing turnips for winter livestock feed |
—
|
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: growing turnips for winter livestock feed | Statement: [Norfolk four-course rotation, componentPractice, growing turnips for winter livestock feed]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: componentPractice Context triple: [Norfolk four-course rotation, componentPractice, growing turnips for winter livestock feed]
-
A.
component3
Indicates that one entity is the third component or sub-part within a larger composite structure or system involving another entity.
-
B.
component1
Indicates that one entity functions as a constituent or part of another entity within a larger whole.
-
C.
componentAct
Indicates that an action or behavior is performed by a component as part of a larger system or process.
-
D.
componentUnit
Indicates that one entity functions as a constituent or sub-unit that forms part of a larger composite entity.
-
E.
componentFrom
Indicates that one entity is derived from, produced by, or originates as a component of another entity.
- 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_69c008cd17c8819082b82d3fbeb68047 |
completed | March 22, 2026, 3:20 p.m. |
| NER | Named-entity recognition | batch_69c063f956c08190ae0f198ccbd68b42 |
completed | March 22, 2026, 9:49 p.m. |
| PD | Predicate disambiguation | batch_69c05608a5608190b22a1fdc4060470d |
completed | March 22, 2026, 8:50 p.m. |
| PDg | Predicate description generation | batch_69c05b37ac1881909e947822490ba4f2 |
completed | March 22, 2026, 9:12 p.m. |
Created at: March 22, 2026, 4:26 p.m.