Triple

T13184109
Position Surface form Disambiguated ID Type / Status
Subject Government Employees Insurance Company E313802 entity
Predicate founder P104 FINISHED
Object Leo Goodwin Sr. E313803 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: Leo Goodwin Sr. | Statement: [Government Employees Insurance Company, founder, Leo Goodwin Sr.]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Leo Goodwin Sr.
Context triple: [Government Employees Insurance Company, founder, Leo Goodwin Sr.]
  • A. Leo Goodwin Sr. chosen
    Leo Goodwin Sr. was an American businessman best known for creating the low-cost auto insurance model that made GEICO one of the largest insurers in the United States.
  • B. Harold Goodwin
    Harold Goodwin was an American character actor active in the early to mid-20th century, known for his numerous supporting roles in both silent and sound films.
  • C. James Haygood
    James Haygood is a film editor known for his work on major feature films and commercials, including editing the thriller "Panic Room."
  • D. Bill Goodwin
    Bill Goodwin was an American radio and television announcer and actor best known for his work on comedy programs in the 1940s and 1950s.
  • E. George Winslow
    George Winslow was an American child actor of the 1950s, best remembered for his deadpan delivery and distinctive raspy voice in several popular Hollywood comedies.
  • 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_69d806ae1e08819090d95bfe1538cc17 completed April 9, 2026, 8:06 p.m.
NER Named-entity recognition batch_69d98c4a0b0081908027bf77442ff5ff completed April 10, 2026, 11:48 p.m.
NED1 Entity disambiguation (via context triple) batch_69f6f5f307408190afd0df16a417c456 completed May 3, 2026, 7:14 a.m.
Created at: April 9, 2026, 9:15 p.m.