Triple
T7775359
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Levinstein |
E221376
|
entity |
| Predicate | hasRelatedSurname |
P3889
|
FINISHED |
| Object | Lowenstein |
E221374
|
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: Lowenstein | Statement: [Levinstein, hasRelatedSurname, Lowenstein]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Lowenstein Context triple: [Levinstein, hasRelatedSurname, Lowenstein]
-
A.
Lowenstein
chosen
Lowenstein is a surname of German origin, often associated with Jewish families and various notable individuals in fields such as politics, academia, and the arts.
-
B.
Bronstein
Bronstein is the original family surname of the Russian revolutionary leader Leon Trotsky.
-
C.
Rosenfeld
Rosenfeld is a notable architectural work designed by 19th-century American architect Alexander Jackson Davis.
-
D.
Greenberg
Greenberg is a 2010 indie dramedy film directed by Noah Baumbach and starring Ben Stiller as a troubled man reevaluating his life while housesitting in Los Angeles.
-
E.
Loebl Schlossman & Hackl
Loebl Schlossman & Hackl is a Chicago-based architectural firm known for designing prominent commercial and high-rise buildings.
- 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_69ca83ebbef881909ac47f789145fef7 |
completed | March 30, 2026, 2:08 p.m. |
| NER | Named-entity recognition | batch_69caa4d005808190ac14c8d716421bdb |
completed | March 30, 2026, 4:29 p.m. |
| NED1 | Entity disambiguation (via context triple) | batch_69cb59ca4c088190bc4f1f9b2488d996 |
completed | March 31, 2026, 5:21 a.m. |
Created at: March 30, 2026, 3:45 p.m.