Triple

T14062172
Position Surface form Disambiguated ID Type / Status
Subject Trinket Island E338371 entity
Predicate naturalDisasterImpact P76614 FINISHED
Object severe damage in 2004 tsunami LITERAL 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: severe damage in 2004 tsunami | Statement: [Trinket Island, naturalDisasterImpact, severe damage in 2004 tsunami]
PD Predicate disambiguation gpt-5-mini-2025-08-07
Target predicate: naturalDisasterImpact
Context triple: [Trinket Island, naturalDisasterImpact, severe damage in 2004 tsunami]
  • A. causeOfDisaster
    Indicates that the subject is responsible for bringing about or triggering the specified disaster.
  • B. disasterDepicted
    Indicates that one entity visually represents or portrays a disaster involving or affecting another entity.
  • C. disasterLocation chosen
    Indicates the place where a disaster occurs or has its primary impact.
  • D. notableDisasterType
    Indicates the specific kind or category of disaster for which something (such as a place, event, or entity) is notable or best known.
  • E. frequentNaturalHazard
    Indicates that a location or area regularly experiences natural hazards such as floods, earthquakes, storms, or similar events with notable frequency.
  • F. None of above.

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_69d81c67ba6c819091935650dfb3b895 completed April 9, 2026, 9:38 p.m.
NER Named-entity recognition batch_69de568876308190840361dcaf10bd45 completed April 14, 2026, 3 p.m.
PD Predicate disambiguation batch_69de05adef888190b023ab42ef5076b6 completed April 14, 2026, 9:15 a.m.
Created at: April 9, 2026, 10:21 p.m.