Triple
T1019786
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | American flamingo |
E22012
|
entity |
| Predicate | feedingBehavior |
P15764
|
FINISHED |
| Object | filter feeding |
—
|
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: filter feeding | Statement: [American flamingo, feedingBehavior, filter feeding]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: feedingBehavior Context triple: [American flamingo, feedingBehavior, filter feeding]
-
A.
feedingStructure
Indicates a relationship where one entity serves as the anatomical or mechanical structure used by another entity to obtain or ingest food.
-
B.
feedingType
chosen
Indicates the manner or method by which one entity provides nourishment or food to another.
-
C.
includesFeedingType
Indicates that one entity encompasses or specifies a particular type or category of feeding associated with another entity.
-
D.
primaryFood
Indicates that one entity serves as the main or most important food source for another entity.
-
E.
preysOn
Indicates that one entity hunts, kills, and consumes another entity as a food source.
- 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_69a493d6e380819097b384986ffc315c |
completed | March 1, 2026, 7:30 p.m. |
| NER | Named-entity recognition | batch_69a4b7dbcf7c8190858b2d16a27bd2ff |
completed | March 1, 2026, 10:04 p.m. |
| PD | Predicate disambiguation | batch_69a4b724c7908190a5b92a57fbdbff4e |
completed | March 1, 2026, 10:01 p.m. |
Created at: March 1, 2026, 7:41 p.m.