Triple
T22129007
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Poodle Hat |
E546860
|
entity |
| Predicate | hasPart |
P35
|
FINISHED |
| Object | eBay |
—
|
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: eBay | Statement: [Poodle Hat, hasPart, eBay]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: eBay Context triple: [Poodle Hat, hasPart, eBay]
-
A.
eBay
chosen
eBay is a global e-commerce company best known for its online auction and consumer-to-consumer sales platform.
-
B.
Cazoo
Cazoo is a UK-based online car retailer known for its high-profile sports sponsorships and rapid growth in the digital automotive marketplace.
-
C.
Alibaba.com
Alibaba.com is a leading global online business-to-business (B2B) marketplace that connects suppliers, primarily manufacturers and wholesalers, with buyers around the world.
-
D.
Yahoo! JAPAN Auctions
Yahoo! JAPAN Auctions is a major Japanese online auction and marketplace platform where users can buy and sell a wide range of goods.
-
E.
Etsy
Etsy is an online marketplace specializing in handmade, vintage, and unique factory-manufactured items, connecting independent sellers with buyers worldwide.
- 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_69e11e39bf348190b541bfa16a7b71e0 |
completed | April 16, 2026, 5:36 p.m. |
| NER | Named-entity recognition | batch_69f12983acfc81908013f66acb31f198 |
completed | April 28, 2026, 9:41 p.m. |
Created at: April 16, 2026, 8:32 p.m.