Triple
T14356398
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Kiki van Eijk |
E355979
|
entity |
| Predicate | hasWorkedFor |
P11675
|
FINISHED |
| Object | Royal Tichelaar Makkum |
E1095456
|
NE 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: Royal Tichelaar Makkum | Statement: [Kiki van Eijk, hasWorkedFor, Royal Tichelaar Makkum]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Royal Tichelaar Makkum Context triple: [Kiki van Eijk, hasWorkedFor, Royal Tichelaar Makkum]
-
A.
Royal Tichelaar Makkum
chosen
Royal Tichelaar Makkum is a historic Dutch ceramics manufacturer renowned for its traditional and contemporary earthenware and tile production.
-
B.
Heer van Mechelen
Heer van Mechelen is the Dutch title historically used for the feudal Lord of the city of Mechelen in present-day Belgium.
-
C.
Theo Heemskerk
Theo Heemskerk was a Dutch politician who served as Prime Minister of the Netherlands in the early 20th century.
-
D.
Boesinghe
Boesinghe is a village in West Flanders, Belgium, known for its proximity to key World War I battlefields along the Yser Front.
-
E.
Hado van Hasselt
Hado van Hasselt is a researcher in reinforcement learning best known for pioneering methods such as Double Q-learning and Dueling DQN that address overestimation bias and improve deep RL performance.
- F. None of above.
- G. Unsure - the case is ambiguous/there is not enough information to decide.
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_69d82790a7e08190877e2d349b2e8d8e |
completed | April 9, 2026, 10:26 p.m. |
| NER | Named-entity recognition | batch_69de8f519bf881908615f4d47e0f77aa |
completed | April 14, 2026, 7:02 p.m. |
| NED1 | Entity disambiguation (via context triple) | batch_69fd550ca6b88190b76cd486bdd66fdf |
completed | May 8, 2026, 3:14 a.m. |
Created at: April 10, 2026, 1:15 a.m.