Triple
T7831083
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Howden Minster |
E181368
|
entity |
| Predicate | locatedIn |
P40
|
FINISHED |
| Object | Howden |
E181368
|
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: Howden | Statement: [Howden Minster, locatedIn, Howden]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Howden Context triple: [Howden Minster, locatedIn, Howden]
-
A.
Howden
Howden is a residential neighbourhood within the town of Livingston in West Lothian, Scotland.
-
B.
Howden
chosen
Howden is a historic market town in the East Riding of Yorkshire, England, known for its medieval Minster and traditional town centre.
-
C.
Hoyt
Hoyt is a family surname of English origin borne by various notable individuals, including Mary Hoyt Sherman before her marriage.
-
D.
Dunvant
Dunvant is a suburban village and community in the City and County of Swansea in south Wales.
-
E.
Southworth
Southworth is a surname most notably associated with Lucinda Southworth, an American researcher and the wife of Google co-founder Larry Page.
- 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_69ca8282ccec819083c48efb72d21cf9 |
completed | March 30, 2026, 2:02 p.m. |
| NER | Named-entity recognition | batch_69cb0647848081908d74e09f52d0919e |
completed | March 30, 2026, 11:24 p.m. |
| NED1 | Entity disambiguation (via context triple) | batch_69cb5a906b908190a204779950555edc |
completed | March 31, 2026, 5:24 a.m. |
Created at: March 30, 2026, 4:44 p.m.