Triple
T236787
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Bengal |
E4840
|
entity |
| Predicate | majorCity |
P316
|
FINISHED |
| Object | Dhaka |
E26021
|
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: Dhaka | Statement: [Bengal, majorCity, Dhaka]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Dhaka Context triple: [Bengal, majorCity, Dhaka]
-
A.
Dhaka
chosen
Dhaka is the capital and largest city of Bangladesh, serving as the country’s political, economic, and cultural center.
-
B.
Calcutta
Calcutta, now known as Kolkata, is a major cultural and commercial metropolis in eastern India that served as the capital of British India until the early 20th century.
-
C.
Guwahati
Guwahati is a major city in northeastern India, serving as a key cultural, economic, and transportation hub for the region.
-
D.
Karachi
Karachi is Pakistan’s sprawling economic hub and major port city on the Arabian Sea, known for its diverse population and central role in the country’s finance, industry, and culture.
-
E.
Islamabad
Islamabad is Pakistan’s planned, modern capital city known for its high standard of living, greenery, and role as the country’s political and administrative center.
- 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_69a257c3d0708190b0871c4269d273e6 |
completed | Feb. 28, 2026, 2:49 a.m. |
| NER | Named-entity recognition | batch_69a25ccc5d548190b505bf1d99db41bd |
completed | Feb. 28, 2026, 3:11 a.m. |
| NED1 | Entity disambiguation (via context triple) | batch_69a3695ec8cc8190a070462cd0022f6a |
completed | Feb. 28, 2026, 10:17 p.m. |
Created at: Feb. 28, 2026, 2:53 a.m.