Triple

T3953976
Position Surface form Disambiguated ID Type / Status
Subject Var E84932 entity
Predicate contains P35 FINISHED
Object Porquerolles E366421 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: Porquerolles | Statement: [Var, contains, Porquerolles]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Porquerolles
Context triple: [Var, contains, Porquerolles]
  • A. Porquerolles chosen
    Porquerolles is a Mediterranean island off the southern coast of France, renowned for its unspoiled beaches, pine forests, and protected natural landscapes within the Îles d'Hyères archipelago.
  • B. Port-Cros
    Port-Cros is a small, protected Mediterranean island off the southern coast of France, renowned for its national park, unspoiled landscapes, and rich marine biodiversity.
  • C. Cap-d’Ail
    Cap-d’Ail is a coastal resort town on the French Riviera, bordering Monaco and known for its scenic beaches and Belle Époque villas.
  • D. Île d’Yeu
    Île d’Yeu is a small French Atlantic island off the Vendée coast, known for its rugged coastline, sandy beaches, and historic fortifications.
  • E. La Digue
    La Digue is a small Seychellois island famed for its granite boulder-fringed beaches, turquoise waters, and laid-back, traditional island atmosphere.
  • 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_69aed934fbfc8190847068e4546de963 completed March 9, 2026, 2:29 p.m.
NER Named-entity recognition batch_69aef93b8f5c8190bdb062a76b68b3e0 completed March 9, 2026, 4:45 p.m.
NED1 Entity disambiguation (via context triple) batch_69b53ffc49f48190949d14113031dd85 completed March 14, 2026, 11:01 a.m.
Created at: March 9, 2026, 3:30 p.m.