Triple
T17243405
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Leoš Janáček |
E418558
|
entity |
| Predicate | placeOfDeath |
P21
|
FINISHED |
| Object | Ostrava |
—
|
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: Ostrava | Statement: [Leoš Janáček, placeOfDeath, Ostrava]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Ostrava Context triple: [Leoš Janáček, placeOfDeath, Ostrava]
-
A.
Ostrava
chosen
Ostrava is a major industrial and cultural city in the northeastern Czech Republic, near the borders with Poland and Slovakia.
-
B.
Opava
Opava is a historic city in the Czech Republic’s Silesian region, known as a former political and cultural center of Silesia.
-
C.
Plzeň
Plzeň is a major city in western Bohemia in the Czech Republic, known for its brewing tradition and industrial heritage.
-
D.
Brno
Brno is the second-largest city in the Czech Republic, known as a major cultural, educational, and industrial center in the historical region of Moravia.
-
E.
Olomouc
Olomouc is a historic city in the eastern Czech Republic known for its well-preserved old town, Baroque architecture, and UNESCO-listed Holy Trinity Column.
- F. None of above.
- G. Unsure - the case is ambiguous/there is not enough information to decide.
Provenance (2 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_69d886d8e96081909870bff6c3d0bf09 |
completed | April 10, 2026, 5:12 a.m. |
| NER | Named-entity recognition | batch_69e42e21bb5c8190ad960f231fe54665 |
completed | April 19, 2026, 1:21 a.m. |
Created at: April 10, 2026, 5:39 a.m.