Triple
T7702811
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Exxon Valdez oil spill |
E174539
|
entity |
| Predicate | affectedSpecies |
P2895
|
FINISHED |
| Object | sea otters |
—
|
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: sea otters | Statement: [Exxon Valdez oil spill, affectedSpecies, sea otters]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: affectedSpecies Context triple: [Exxon Valdez oil spill, affectedSpecies, sea otters]
-
A.
someSpeciesAre
Indicates that at least one member of a specified group or category belongs to, or can be classified as, a particular species.
-
B.
laterSpecies
Indicates that one species appears or evolves later in time relative to another species.
-
C.
affectsTaxon
chosen
Indicates that one entity has an impact or influence on a particular taxon or group of organisms.
-
D.
includesSpeciesCausing
Indicates that one entity contains or encompasses species that are responsible for causing a particular effect, condition, or outcome in another entity.
-
E.
featuresSpecies
Indicates that something includes, presents, or highlights a particular species as part of its content or composition.
- 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_69c6995a72cc8190998e56daa6f8e453 |
completed | March 27, 2026, 2:51 p.m. |
| NER | Named-entity recognition | batch_69c70402169481909b219dc5f4a64b9b |
completed | March 27, 2026, 10:26 p.m. |
| PD | Predicate disambiguation | batch_69c70165e78c8190bf6b3c34e243cb81 |
completed | March 27, 2026, 10:15 p.m. |
Created at: March 27, 2026, 4:03 p.m.