Triple
T20824049
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Baltic Sea islands |
E512649
|
entity |
| Predicate | includes |
P1393
|
FINISHED |
| Object | Saaremaa |
—
|
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: Saaremaa | Statement: [Baltic Sea islands, includes, Saaremaa]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Saaremaa Context triple: [Baltic Sea islands, includes, Saaremaa]
-
A.
Saaremaa
chosen
Saaremaa is the largest island of Estonia, known for its rugged coastline, medieval Kuressaare Castle, and distinctive windmills and juniper landscapes.
-
B.
Naissaar
Naissaar is a forested Estonian island in the Baltic Sea known for its military history, nature reserves, and proximity to Tallinn.
-
C.
Hiiumaa
Hiiumaa is Estonia’s second-largest island, located in the Baltic Sea and known for its unspoiled nature, lighthouses, and quiet rural landscapes.
-
D.
Märjamaa
Märjamaa is a small borough in western Estonia that serves as a local administrative and service center within Rapla County.
-
E.
Kuressaare
Kuressaare is the main town on Estonia’s Saaremaa island, known for its well-preserved medieval castle and seaside spa resort atmosphere.
- 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_69e0b4ce39108190a6e8e5df4f1c8dc5 |
completed | April 16, 2026, 10:07 a.m. |
| NER | Named-entity recognition | batch_69e6c2fc0cd081909e264cda686579ea |
completed | April 21, 2026, 12:21 a.m. |
Created at: April 16, 2026, 12:41 p.m.