Triple

T11676303
Position Surface form Disambiguated ID Type / Status
Subject Keikyu Corporation E277499 entity
Predicate shortName P43 FINISHED
Object Keikyu E548853 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: Keikyu | Statement: [Keikyu Corporation, shortName, Keikyu]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Keikyu
Context triple: [Keikyu Corporation, shortName, Keikyu]
  • A. Miyakejima
    Miyakejima is a volcanic island in the Izu archipelago of Japan, known for its active volcano, frequent eruptions, and unique gas-monitoring measures affecting residents and visitors.
  • B. Keikyū chosen
    Keikyū is a major private railway operator in the Greater Tokyo area of Japan, known for its commuter and airport rail services.
  • C. Jinoshima
    Jinoshima is one of the uninhabited Tomogashima Islands in Japan, known for its rugged natural scenery and remnants of former military fortifications.
  • D. Sannō-jima
    Sannō-jima is a small Japanese island that forms part of the remote Mukojima subgroup within the Ogasawara (Bonin) Islands in the Pacific Ocean.
  • E. Yome-jima
    Yome-jima is a small uninhabited island that is part of the Ogasawara (Bonin) Islands chain in the Pacific Ocean, known for its rugged coastline and protected natural environment.
  • 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_69d6aafd0a448190b44da30af8c6c519 completed April 8, 2026, 7:22 p.m.
NER Named-entity recognition batch_69d8a44504c48190b519765a83ff9c5e completed April 10, 2026, 7:18 a.m.
NED1 Entity disambiguation (via context triple) batch_69fe6b366470819093a74828e2a85116 completed May 8, 2026, 11:01 p.m.
Created at: April 8, 2026, 9:40 p.m.