Triple
T35685777
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Emily (The Devil Wears Prada) |
E1031142
|
entity |
| Predicate | losesOpportunityTo |
P157426
|
FINISHED |
| Object | Andy Sachs |
—
|
NE NERFINISHED |
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: Andy Sachs | Statement: [Emily (The Devil Wears Prada), losesOpportunityTo, Andy Sachs]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: losesOpportunityTo Context triple: [Emily (The Devil Wears Prada), losesOpportunityTo, Andy Sachs]
-
A.
offersOpportunity
Indicates that one entity provides another with a chance or possibility to do, obtain, or experience something.
-
B.
promotionPossibility
Indicates the potential or eligibility for an entity to be advanced to a higher rank, position, or status.
-
C.
lostOpportunity
chosen
Indicates that a potential benefit, advantage, or desired outcome that could have been achieved through an action or relationship was missed or not realized.
-
D.
offersConnectionTo
Indicates that one entity provides or enables access, linkage, or connectivity to another entity.
-
E.
offersExchangeOpportunities
Indicates that one entity provides chances for another entity to participate in exchanges, such as trading, swapping, or reciprocal programs.
- 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_69f76e0bb6608190ad3a1880be54a17d |
completed | May 3, 2026, 3:47 p.m. |
| NER | Named-entity recognition | batch_69f7a01fadac8190a429777cec614eca |
completed | May 3, 2026, 7:21 p.m. |
| PD | Predicate disambiguation | batch_69f79e4d885881908a3612e2e75cf84f |
completed | May 3, 2026, 7:13 p.m. |
Created at: May 3, 2026, 4:05 p.m.