Triple

T13375103
Position Surface form Disambiguated ID Type / Status
Subject Helga E319163 entity
Predicate relatedToName P3889 FINISHED
Object Olga E136344 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: Olga | Statement: [Helga, relatedToName, Olga]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Olga
Context triple: [Helga, relatedToName, Olga]
  • A. Olga chosen
    Olga is a female given name of Russian origin, historically borne by several notable figures including Russian grand duchesses and saints.
  • B. Olga
    Olga is a lingerie and intimate apparel brand owned by PVH Corp., known for designing comfortable, supportive undergarments for women.
  • C. Galina
    Galina is a feminine given name of Slavic origin, commonly used in Russia and other Eastern European countries.
  • D. Gavriella
    Gavriella is a feminine given name, typically considered a variant of Gabriella with similar Hebrew and Italian roots meaning "God is my strength."
  • E. Olga Lysova
    Olga Lysova is a Russian woman best known as the first wife of billionaire businessman and former Chelsea F.C. owner Roman Abramovich.
  • 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_69d806b886bc8190b676e7768b8e01c5 completed April 9, 2026, 8:06 p.m.
NER Named-entity recognition batch_69dadcda64a48190b53243a763cd175b completed April 11, 2026, 11:44 p.m.
NED1 Entity disambiguation (via context triple) batch_69f7461814dc8190aaaefc648da75246 completed May 3, 2026, 12:56 p.m.
Created at: April 9, 2026, 9:33 p.m.