Triple
T19366754
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Lezayre |
E484421
|
entity |
| Predicate | borders |
P224
|
FINISHED |
| Object | Andreas |
—
|
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: Andreas | Statement: [Lezayre, borders, Andreas]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Andreas Context triple: [Lezayre, borders, Andreas]
-
A.
Andreas
chosen
Andreas is a village on the Isle of Man, known for its rural character and historic parish church.
-
B.
Andreas
Andreas is a masculine given name of Greek origin, commonly used in various European and international cultures.
-
C.
Johan
Johan is the given first name of the Swedish playwright and novelist August Strindberg.
-
D.
Johan
Johan is a masculine given name of Scandinavian origin, commonly used in countries such as Norway, Sweden, and Denmark.
-
E.
Johan
Johan is the given first name of J. Erik Jonsson, an American businessman and philanthropist who co-founded Texas Instruments and served as mayor of Dallas.
- 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_69d8e8d305088190ad13571532aa454c |
completed | April 10, 2026, 12:10 p.m. |
| NER | Named-entity recognition | batch_69e619ac26d4819095836d737b629cf1 |
completed | April 20, 2026, 12:18 p.m. |
Created at: April 10, 2026, 1:35 p.m.