Triple
T23060167
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Bunung |
E574275
|
entity |
| Predicate | hasDialect |
P4251
|
FINISHED |
| Object | Takbanuaz dialect |
—
|
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: Takbanuaz dialect | Statement: [Bunung, hasDialect, Takbanuaz dialect]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Takbanuaz dialect Context triple: [Bunung, hasDialect, Takbanuaz dialect]
-
A.
Takbanuaz dialect
chosen
The Takbanuaz dialect is a regional variety of the Bunun language spoken by an indigenous Bunun subgroup in Taiwan.
-
B.
Zanniat dialect
The Zanniat dialect is a regional variety of the Falam Chin language spoken by the Zanniat people in parts of Chin State, Myanmar.
-
C.
Kebkabiya dialect
The Kebkabiya dialect is a regional variety of the Foor language spoken around the Kebkabiya area, distinguished by its local phonological and lexical features.
-
D.
Bakhah dialect
The Bakhah dialect is a regional variety of Arabic spoken in parts of Syria, closely associated with the speech community of Bakhah village and sharing many features with neighboring mountain dialects.
-
E.
Razihi dialect
The Razihi dialect is a highly distinctive and conservative Arabic variety spoken in parts of northwestern Yemen, noted for preserving many archaic linguistic features.
- 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_69e245ba7ae48190be606dbc54120e39 |
completed | April 17, 2026, 2:37 p.m. |
| NER | Named-entity recognition | batch_69f1899f359081909e89e19db3833a3d |
completed | April 29, 2026, 4:31 a.m. |
Created at: April 17, 2026, 3:55 p.m.