Triple

T4446792
Position Surface form Disambiguated ID Type / Status
Subject Alma Reville E96307 entity
Predicate child P120 FINISHED
Object Patricia Hitchcock E102454 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: Patricia Hitchcock | Statement: [Alma Reville, child, Patricia Hitchcock]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Patricia Hitchcock
Context triple: [Alma Reville, child, Patricia Hitchcock]
  • A. Patricia Hitchcock chosen
    Patricia Hitchcock was a British-American actress best known for her supporting roles in several of her father Alfred Hitchcock’s films and appearances on his television series.
  • B. Alma Reville
    Alma Reville was a British screenwriter, editor, and assistant director best known for her long creative collaboration with and marriage to filmmaker Alfred Hitchcock.
  • C. Johanna Mankiewicz Davis
    Johanna Mankiewicz Davis was an American novelist and screenwriter, and the daughter of acclaimed screenwriter Herman J. Mankiewicz.
  • D. Ruth Wells
    Ruth Wells was the wife of three-time Academy Award–winning American character actor Walter Brennan.
  • E. Priscilla Lane
    Priscilla Lane was an American film actress best known as one of the Lane Sisters and for her roles in popular 1930s and 1940s Hollywood films.
  • 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_69b345415ba481908df738e7174448ba completed March 12, 2026, 10:59 p.m.
NER Named-entity recognition batch_69b355d31e10819086590b9f828d50b0 completed March 13, 2026, 12:09 a.m.
NED1 Entity disambiguation (via context triple) batch_69b6376286e48190906874d67c730dc9 completed March 15, 2026, 4:36 a.m.
Created at: March 12, 2026, 11:32 p.m.