Triple
T30635283
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Jason Gideon |
E779824
|
entity |
| Predicate | leavesOrganization |
P2983
|
FINISHED |
| Object | Behavioral Analysis Unit |
—
|
NE NERFINISHED |
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: Behavioral Analysis Unit | Statement: [Jason Gideon, leavesOrganization, Behavioral Analysis Unit]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: leavesOrganization Context triple: [Jason Gideon, leavesOrganization, Behavioral Analysis Unit]
-
A.
leavesEmployer
Indicates that an employee ends or terminates their working relationship with an employer.
-
B.
leavesCommunity
Indicates that an entity departs from or is no longer a member of a particular community.
-
C.
leftOrganization
chosen
Indicates that an entity has departed from or is no longer affiliated with a particular organization.
-
D.
endedOrganization
Indicates that an organization has ceased to exist or operate, marking the termination of its existence or activities.
-
E.
entryOrganization
Indicates the organization responsible for creating, submitting, or maintaining the associated entry.
- F. None of above.
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_69f224a50ebc81909b961a94c7f66b12 |
completed | April 29, 2026, 3:32 p.m. |
| NER | Named-entity recognition | batch_69f68a5062f08190be313f1baa1a317d |
completed | May 2, 2026, 11:35 p.m. |
| PD | Predicate disambiguation | batch_69f67e448a9c8190b591374d98799fe3 |
completed | May 2, 2026, 10:44 p.m. |
Created at: April 29, 2026, 8:28 p.m.