Triple

T14565487
Position Surface form Disambiguated ID Type / Status
Subject 東沙環礁 E341773 entity
Predicate 別名 P39 FINISHED
Object Pratas Islands E68524 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: Pratas Islands | Statement: [東沙環礁, 別名, Pratas Islands]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Pratas Islands
Context triple: [東沙環礁, 別名, Pratas Islands]
  • A. Pratas Islands chosen
    The Pratas Islands are a small, strategically located atoll in the northern South China Sea administered by Taiwan but also claimed by China.
  • B. Subi Islands
    The Subi Islands are a small island group within Indonesia’s Natuna archipelago in the South China Sea, known for their strategic maritime location and rich surrounding marine biodiversity.
  • C. Tayando Islands
    The Tayando Islands are a small archipelago in Indonesia’s Maluku province, known for their remote location, coral-rich seas, and traditional island communities.
  • D. Trang Islands
    The Trang Islands are a group of picturesque, relatively undeveloped islands in the Andaman Sea known for their white-sand beaches, limestone cliffs, and rich marine life.
  • E. Wala Islands
    Wala Islands are a small group of islands in northern Vanuatu known for their local communities and use of the Uripiv-Wala-Rano-Atchin language.
  • 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_69d822dcc6248190bed689984bceb0e2 completed April 9, 2026, 10:06 p.m.
NER Named-entity recognition batch_69deb38c6350819091090ffd15772f6f completed April 14, 2026, 9:37 p.m.
NED1 Entity disambiguation (via context triple) batch_69ff56afa5ec8190a058574dff7431dc completed May 9, 2026, 3:45 p.m.
Created at: April 10, 2026, 1:23 a.m.