Triple

T14381407
Position Surface form Disambiguated ID Type / Status
Subject Kadavu Province E356610 entity
Predicate administrativeCenter P1474 FINISHED
Object Vunisea E352339 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: Vunisea | Statement: [Kadavu Province, administrativeCenter, Vunisea]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Vunisea
Context triple: [Kadavu Province, administrativeCenter, Vunisea]
  • A. Vunisea chosen
    Vunisea is a small coastal village that serves as the administrative and commercial center of Kadavu Island in Fiji.
  • B. Merelava
    Merelava is a small volcanic island in northern Vanuatu, known for its steep terrain, active cultural traditions, and use of the Mwerlap language by its inhabitants.
  • C. Nawuli
    Nawuli is an alternative name for the Nawuri language, a Guang language spoken primarily in parts of Ghana.
  • D. Vaiaku
    Vaiaku is a small village on the atoll of Funafuti in Tuvalu that serves as the country's administrative center.
  • E. Muribenua
    Muribenua is a village on the low-lying coral atoll of Nikunau in the Republic of Kiribati, a Pacific island nation.
  • 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_69d8279163a081908aec45c0e3f1e02f completed April 9, 2026, 10:26 p.m.
NER Named-entity recognition batch_69de900bbfb08190a1e56f281a2374c0 completed April 14, 2026, 7:05 p.m.
NED1 Entity disambiguation (via context triple) batch_69fd5bc0c9708190b5025e9675e0e925 completed May 8, 2026, 3:42 a.m.
Created at: April 10, 2026, 1:16 a.m.