Triple
T32187701
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Discover Card |
E822150
|
entity |
| Predicate | cardholderBenefits |
P2188
|
FINISHED |
| Object | cash-back on purchases |
—
|
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: cash-back on purchases | Statement: [Discover Card, cardholderBenefits, cash-back on purchases]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: cardholderBenefits Context triple: [Discover Card, cardholderBenefits, cash-back on purchases]
-
A.
cardEligibility
Indicates whether an entity qualifies for or is allowed to receive a particular card under specified criteria.
-
B.
hasBenefit
chosen
Indicates that one entity provides an advantage, improvement, or positive outcome to another entity.
-
C.
benefitAppliesTo
Indicates that a particular benefit is applicable to, or valid for, a specified entity or context.
-
D.
benefitsAre
Indicates that certain advantages, gains, or positive outcomes are possessed by or accrue to a particular entity or group.
-
E.
benefitCharacteristic
Indicates that one entity possesses a quality or feature that provides an advantage, usefulness, or positive effect to another entity.
- 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_69f3490819cc81909bae1f8ce99423c5 |
completed | April 30, 2026, 12:20 p.m. |
| NER | Named-entity recognition | batch_69f6babc938c8190a233c3a8b5254802 |
completed | May 3, 2026, 3:02 a.m. |
| PD | Predicate disambiguation | batch_69f6b3aa892481908d29283a074e6722 |
completed | May 3, 2026, 2:32 a.m. |
Created at: May 1, 2026, 12:35 a.m.