Triple
T22662772
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Actimel |
E559705
|
entity |
| Predicate | hasVariant |
P455
|
FINISHED |
| Object | no-added-sugar Actimel |
—
|
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: no-added-sugar Actimel | Statement: [Actimel, hasVariant, no-added-sugar Actimel]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: no-added-sugar Actimel Context triple: [Actimel, hasVariant, no-added-sugar Actimel]
-
A.
Actimel
chosen
Actimel is a probiotic yogurt drink brand known for its small bottled dairy shots marketed to support the immune system.
-
B.
Activia
Activia is a popular probiotic yogurt brand known for supporting digestive health, produced and marketed by Danone.
-
C.
Sprite Zero Sugar Lymonade
Sprite Zero Sugar Lymonade is a no-sugar, lemon-lime flavored soft drink that offers the taste of Sprite Lymonade with zero calories.
-
D.
Yakult Atoms
Yakult Atoms was the former name of the professional Japanese baseball team now known as the Tokyo Yakult Swallows.
-
E.
Milk
Milk is a nutrient-rich white liquid produced by the mammary glands of mammals and widely consumed as a staple food and beverage by humans.
- F. None of above.
- G. Unsure - the case is ambiguous/there is not enough information to decide.
Provenance (2 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_69e2454a158c819093b8e35f5045efb6 |
completed | April 17, 2026, 2:35 p.m. |
| NER | Named-entity recognition | batch_69f17660c0c88190bed9fa8f6517eec4 |
completed | April 29, 2026, 3:09 a.m. |
Created at: April 17, 2026, 3:08 p.m.