Triple

T6188368
Position Surface form Disambiguated ID Type / Status
Subject Lesser Sunda Islands E138121 entity
Predicate hasPart P35 FINISHED
Object Adonara E444589 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: Adonara | Statement: [Lesser Sunda Islands, hasPart, Adonara]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Adonara
Context triple: [Lesser Sunda Islands, hasPart, Adonara]
  • A. Adonara Island chosen
    Adonara Island is a small volcanic island in eastern Indonesia’s Lesser Sunda Islands, known for its traditional communities and location between Flores and Lembata.
  • B. Hatoma Island
    Hatoma Island is a small, remote island in Japan’s Yaeyama archipelago, known for its traditional Ryukyuan culture, coral reefs, and tranquil, sparsely populated environment.
  • C. Uncatena Island
    Uncatena Island is a small, privately owned island in the Elizabeth Islands chain off the coast of Massachusetts, known for its limited access and natural, undeveloped character.
  • D. Atauro
    Atauro is a rugged, sparsely populated island north of Dili in Timor-Leste, known for its rich marine biodiversity and excellent diving and snorkeling sites.
  • E. Landana
    Landana is a coastal town in Angola’s Cabinda exclave, historically known as a regional trading and missionary center.
  • 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_69c008a8fd408190b7ec6e42934974a6 completed March 22, 2026, 3:20 p.m.
NER Named-entity recognition batch_69c06217be288190997032c8c4839fa5 completed March 22, 2026, 9:41 p.m.
NED1 Entity disambiguation (via context triple) batch_69c16f06c770819087e055cfe6c8b134 completed March 23, 2026, 4:49 p.m.
Created at: March 22, 2026, 4:19 p.m.