Triple
T22593466
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Mark Levin |
E565008
|
entity |
| Predicate | employer |
P7
|
FINISHED |
| Object | Westwood One |
E10953
|
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: Westwood One | Statement: [Mark Levin, employer, Westwood One]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Westwood One Context triple: [Mark Levin, employer, Westwood One]
-
A.
Westwood One
chosen
Westwood One is a major American radio network best known for its nationwide sports coverage, including live broadcasts of NFL games.
-
B.
Cumulus Media
Cumulus Media is a major American broadcasting company that owns and operates numerous radio stations across the United States.
-
C.
Audacy, Inc.
Audacy, Inc. is a major American multi-platform audio and radio broadcasting company that owns and operates numerous local stations and digital audio brands across the United States.
-
D.
Beasley Broadcast Group
Beasley Broadcast Group is an American media company that owns and operates numerous radio stations across various markets in the United States.
-
E.
iHeartMedia
iHeartMedia is a major American mass media and entertainment company best known for its extensive portfolio of radio stations, digital audio platforms, and live events.
- 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_69e245836014819091b91ed3074742a3 |
completed | April 17, 2026, 2:36 p.m. |
| NER | Named-entity recognition | batch_69f16163cb248190b377b110d80a6730 |
completed | April 29, 2026, 1:39 a.m. |
| NED1 | Entity disambiguation (via context triple) | batch_6a0b3d5e31308190a11bff9270795efd |
completed | May 18, 2026, 4:25 p.m. |
Created at: April 17, 2026, 2:49 p.m.