Triple
T37184305
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Pretoria Accord on the Democratic Republic of the Congo |
E921280
|
entity |
| Predicate | hostCountryOfTalks |
P98941
|
FINISHED |
| Object | South Africa |
—
|
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: South Africa | Statement: [Pretoria Accord on the Democratic Republic of the Congo, hostCountryOfTalks, South Africa]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: hostCountryOfTalks Context triple: [Pretoria Accord on the Democratic Republic of the Congo, hostCountryOfTalks, South Africa]
-
A.
hostCountryOfContest
Indicates that a country serves as the location where a particular contest or competition is held.
-
B.
countryOfSpeaker
Indicates that a person or group is from, associated with, or represents a particular country as a speaker.
-
C.
hostedByCountry
chosen
Indicates that an event or activity is organized, held, or officially hosted within the jurisdiction of a specific country.
-
D.
coHostCountryWith
Indicates that two countries jointly serve as hosts for the same event or activity.
-
E.
mainHostCountry
Indicates the country that primarily serves as the host location for the associated entity or event.
- F. None of above.
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_69f76ea250bc819083f28d81de25cd0c |
completed | May 3, 2026, 3:49 p.m. |
| NER | Named-entity recognition | batch_69fba78aca4c8190b8f1831e8cc04e06 |
completed | May 6, 2026, 8:41 p.m. |
| PD | Predicate disambiguation | batch_69fba34a65a4819088bac6c17542d71c |
completed | May 6, 2026, 8:23 p.m. |
Created at: May 3, 2026, 4:15 p.m.