Triple
T19799708
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Wankie Game Reserve |
E475638
|
entity |
| Predicate | near |
P350
|
FINISHED |
| Object | Victoria Falls |
—
|
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: Victoria Falls | Statement: [Wankie Game Reserve, near, Victoria Falls]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Victoria Falls Context triple: [Wankie Game Reserve, near, Victoria Falls]
-
A.
Victoria Falls
chosen
Victoria Falls is one of the world’s largest and most famous waterfalls, spanning the Zambezi River between Zambia and Zimbabwe and renowned for its dramatic curtain of falling water and mist.
-
B.
Kongo Falls
Kongo Falls is a Super Smash Bros. series fighting stage set on a waterfall-themed section of the Donkey Kong–inspired Kongo Jungle.
-
C.
Tugela Falls
Tugela Falls is a spectacular multi-drop waterfall in the Drakensberg Mountains of South Africa, often cited as one of the tallest waterfalls in the world.
-
D.
Kwa Falls
Kwa Falls is a scenic waterfall and popular natural tourist attraction located within Nigeria’s Cross River State.
-
E.
Kongou Falls
Kongou Falls is a spectacular multi-tiered waterfall on the Ivindo River in Gabon, renowned for its powerful cascades and pristine rainforest surroundings.
- 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_69d8e51bc4208190a1c57d8c5d1b15e4 |
completed | April 10, 2026, 11:55 a.m. |
| NER | Named-entity recognition | batch_69e653cb865c81909696d2b37476f62f |
completed | April 20, 2026, 4:26 p.m. |
Created at: April 10, 2026, 1:49 p.m.