Triple
T18185587
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Hermann Kafka |
E435403
|
entity |
| Predicate | placeOfBirth |
P1
|
FINISHED |
| Object | Wossek |
—
|
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: Wossek | Statement: [Hermann Kafka, placeOfBirth, Wossek]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Wossek Context triple: [Hermann Kafka, placeOfBirth, Wossek]
-
A.
Wossek
chosen
Wossek is a small town in what is now the Czech Republic, historically part of the Austro-Hungarian Empire and known as the birthplace of Hermann Kafka, father of writer Franz Kafka.
-
B.
Wustrow
Wustrow is a small town in the Wendland region of Lower Saxony, Germany, known for its rural character and traditional half-timbered architecture.
-
C.
Wurtulla
Wurtulla is a coastal suburb on Queensland’s Sunshine Coast in Australia, known for its long sandy beaches and residential communities.
-
D.
Wallot
Wallot is a German surname most notably associated with architect Paul Wallot, designer of the Reichstag building in Berlin.
-
E.
Wassenberg
Wassenberg is a historic town in western Germany near the Dutch border, known for its medieval origins and association with the noble House of Wassenberg.
- 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_69d8b90c7ec081909b4694ccecb449c6 |
completed | April 10, 2026, 8:47 a.m. |
| NER | Named-entity recognition | batch_69e4dffdccd881908da772db78b9d081 |
completed | April 19, 2026, 2 p.m. |
Created at: April 10, 2026, 10:31 a.m.