Triple

T4544122
Position Surface form Disambiguated ID Type / Status
Subject Iris E110006 entity
Predicate cinematographyBy P1953 FINISHED
Object Roger Pratt E335683 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: Roger Pratt | Statement: [Iris, cinematographyBy, Roger Pratt]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Roger Pratt
Context triple: [Iris, cinematographyBy, Roger Pratt]
  • A. Roger Pratt chosen
    Roger Pratt is a British cinematographer known for his work on major films such as "Brazil," "Batman," and several entries in the "Harry Potter" series.
  • B. Peter Pratt
    Peter Pratt was a British actor best known to Doctor Who fans for his chilling portrayal of the villainous Time Lord known as the Master.
  • C. Grahame Pratt
    Grahame Pratt is an Australian-born actor and producer best known as the longtime husband and manager of American singer and actress Leslie Uggams.
  • D. Richard Bristow
    Richard Bristow was a 16th-century English Catholic scholar and theologian who contributed to the development and annotation of the Douay–Rheims Bible.
  • E. Richard Hiscott
    Richard Hiscott is an editor known for his work on the television series "Willow."
  • 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_69bd4412524c8190be5bcc9ddee91848 completed March 20, 2026, 12:56 p.m.
NER Named-entity recognition batch_69bd57d517e881909c3d23ed4453b0a7 completed March 20, 2026, 2:21 p.m.
NED1 Entity disambiguation (via context triple) batch_69bdfa1681708190ae4dc66e8734e5a1 completed March 21, 2026, 1:53 a.m.
Created at: March 20, 2026, 1:05 p.m.