Triple

T10408047
Position Surface form Disambiguated ID Type / Status
Subject Inner Banda Arc E245314 entity
Predicate hasIsland P970 FINISHED
Object Romang
Romang is a remote volcanic island in Indonesia’s Maluku province, known for its rugged terrain, small coastal settlements, and location within the tectonically active Banda Sea region.
E861917 NE FINISHED

How this triple was built (4 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: Romang | Statement: [Inner Banda Arc, hasIsland, Romang]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Romang
Context triple: [Inner Banda Arc, hasIsland, Romang]
  • A. Nerekhta
    Nerekhta is a historic Russian town known for its well-preserved traditional architecture and cultural heritage within Kostroma Oblast.
  • B. Rumâri
    Rumâri are an Eastern Romance-speaking ethnic group in Istria, traditionally known as the local Vlach or Istro-Romanian community.
  • C. Lezgin
    Lezgin is a Northeast Caucasian language spoken primarily by the Lezgin people in southern Dagestan (Russia) and northern Azerbaijan.
  • D. Romm
    Romm is a surname most notably associated with Soviet film director and screenwriter Mikhail Romm.
  • E. Rometty
    Rometty is the surname of Ginni Rometty, the former chairman, president, and CEO of IBM.
  • F. None of above. chosen
  • G. Unsure - the case is ambiguous/there is not enough information to decide.
NEDg Description generation gpt-5.1
Instruction
Generate a one-sentence description of the target entity. 
You are given a context triple in the form (subject, predicate, object), where the object is the target entity. 
# Instructions
Use the triple to infer relevant information about the entity. Describe the entity based on what is most defining, well-known. 
Avoid repeating the information from the triple, unless really essential.
# Response Format
Return only the sentence: "Description: [one-sentence description of the target entity]"
Input
Entity: Romang
Triple: [Inner Banda Arc, hasIsland, Romang]
Generated description
Romang is a remote volcanic island in Indonesia’s Maluku province, known for its rugged terrain, small coastal settlements, and location within the tectonically active Banda Sea region.
NED2 Entity disambiguation (via description) gpt-5-mini-2025-08-07
Target entity: Romang
Target entity description: Romang is a remote volcanic island in Indonesia’s Maluku province, known for its rugged terrain, small coastal settlements, and location within the tectonically active Banda Sea region.
  • A. Nerekhta
    Nerekhta is a historic Russian town known for its well-preserved traditional architecture and cultural heritage within Kostroma Oblast.
  • B. Rumâri
    Rumâri are an Eastern Romance-speaking ethnic group in Istria, traditionally known as the local Vlach or Istro-Romanian community.
  • C. Lezgin
    Lezgin is a Northeast Caucasian language spoken primarily by the Lezgin people in southern Dagestan (Russia) and northern Azerbaijan.
  • D. Romm
    Romm is a surname most notably associated with Soviet film director and screenwriter Mikhail Romm.
  • E. Rometty
    Rometty is the surname of Ginni Rometty, the former chairman, president, and CEO of IBM.
  • F. None of above. chosen

Provenance (5 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_69d381be340c8190b05998703d42d224 completed April 6, 2026, 9:49 a.m.
NER Named-entity recognition batch_69d4e9f9b59881909797646fd8cd6a7e completed April 7, 2026, 11:26 a.m.
NED1 Entity disambiguation (via context triple) batch_69d7fbf1c428819099ca359309c4c836 completed April 9, 2026, 7:20 p.m.
NEDg Description generation batch_69d822d597088190bf3dca85e1ddb890 completed April 9, 2026, 10:06 p.m.
NED2 Entity disambiguation (via description) batch_69d859cc1aac8190ab232bb4e4e4fac1 completed April 10, 2026, 2 a.m.
Created at: April 6, 2026, 12:09 p.m.