Triple
T20920095
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Iki Airport |
E515182
|
entity |
| Predicate | pushpinLabel |
P9248
|
FINISHED |
| Object | RJDB |
—
|
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: RJDB | Statement: [Iki Airport, pushpinLabel, RJDB]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: RJDB Context triple: [Iki Airport, pushpinLabel, RJDB]
-
A.
RJDB
chosen
RJDB is the four-letter ICAO airport code assigned to Iki Airport in Japan.
-
B.
RDB
RDB is Rwanda’s government agency responsible for promoting investment, tourism, and economic development in the country.
-
C.
jdb
jdb is the command-line debugger for Java programs that comes bundled with the Oracle JDK.
-
D.
ERDB
ERDB is a Philippine government bureau focused on conducting research and development to support the sustainable management and conservation of the country’s ecosystems and natural resources.
-
E.
DBM
DBM is a demand-side platform from Google’s DoubleClick suite that enables advertisers to buy and manage programmatic display, video, and mobile ad inventory across multiple exchanges.
- 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_69e0b4f9d5ec8190bb2bd27350ed341c |
completed | April 16, 2026, 10:07 a.m. |
| NER | Named-entity recognition | batch_69e6ec677338819081410cbaa2846260 |
completed | April 21, 2026, 3:17 a.m. |
Created at: April 16, 2026, 12:48 p.m.