Triple

T12564167
Position Surface form Disambiguated ID Type / Status
Subject Havas advertising and marketing operations E295424 entity
Predicate partOf P40 FINISHED
Object Havas Group E295423 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: Havas Group | Statement: [Havas advertising and marketing operations, partOf, Havas Group]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Havas Group
Context triple: [Havas advertising and marketing operations, partOf, Havas Group]
  • A. Havas chosen
    Havas is a major global advertising and communications group headquartered in France, known for its extensive network of creative, media, and marketing agencies.
  • B. Publicis Groupe
    Publicis Groupe is a major French multinational advertising and communications company and one of the world’s largest marketing services groups.
  • C. Publicis Communications
    Publicis Communications is a major creative and advertising network within the global marketing and communications conglomerate Publicis Groupe.
  • D. WPP plc
    WPP plc is a British multinational advertising and public relations conglomerate and one of the world’s largest communications groups.
  • E. Omnicom Group
    Omnicom Group is a global marketing and communications holding company that owns numerous leading advertising, media, and public relations agencies worldwide.
  • 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_69d6ad9cac2c81908e8a7bed82d1e21d completed April 8, 2026, 7:33 p.m.
NER Named-entity recognition batch_69d95494ae1c81908b9ee14b8ef92a65 completed April 10, 2026, 7:50 p.m.
NED1 Entity disambiguation (via context triple) batch_69f67c6c21348190b851fce31df307e2 completed May 2, 2026, 10:36 p.m.
Created at: April 8, 2026, 11:49 p.m.