Triple

T8920487
Position Surface form Disambiguated ID Type / Status
Subject Drayton E212401 entity
Predicate hasPostTown P2711 FINISHED
Object Abingdon E38286 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: Abingdon | Statement: [Drayton, hasPostTown, Abingdon]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Abingdon
Context triple: [Drayton, hasPostTown, Abingdon]
  • A. Abingdon chosen
    Abingdon is a historic market town in Oxfordshire, England, often claimed to be one of the oldest continuously inhabited towns in the country.
  • B. Berkhamsted
    Berkhamsted is a historic market town in Hertfordshire, England, known for its medieval castle remains and role as a commuter hub northwest of London.
  • C. Wallingford
    Wallingford is a historic market town on the River Thames in South Oxfordshire, England, known for its medieval castle remains and traditional town centre.
  • D. Wallingford
    Wallingford is a town in south-central Connecticut known for its historic downtown, residential neighborhoods, and mix of light industry and commerce.
  • E. Hamstead
    Hamstead is a suburban residential area in the West Midlands, England, situated between Birmingham and Sandwell.
  • 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_69ca839481d48190b42b037e0d0f636c completed March 30, 2026, 2:07 p.m.
NER Named-entity recognition batch_69cc664f099c8190be6d20a4574212ca completed April 1, 2026, 12:26 a.m.
NED1 Entity disambiguation (via context triple) batch_69d065b51dd881908a7cb8c02562801b completed April 4, 2026, 1:13 a.m.
Created at: March 30, 2026, 6:56 p.m.