Triple
T10471798
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | A86 |
E246939
|
entity |
| Predicate | hasSection |
P35
|
FINISHED |
| Object | A86 duplex tunnel |
E246940
|
NE FINISHED |
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: A86 duplex tunnel | Statement: [A86, hasSection, A86 duplex tunnel]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: A86 duplex tunnel Context triple: [A86, hasSection, A86 duplex tunnel]
-
A.
Duplex A86 tunnel
chosen
The Duplex A86 tunnel is a major double-deck road tunnel in the western suburbs of Paris, designed to complete the A86 ring road while minimizing surface impact.
-
B.
A86
A86 is a major orbital motorway forming part of the ring road system around Paris, France.
-
C.
Tunnel No. 2
Tunnel No. 2 is one of New York City’s major water distribution tunnels that helps convey drinking water from upstate reservoirs to the city’s boroughs.
-
D.
TUNAIR
TUNAIR is the radio callsign used by Tunisair, the national flag carrier airline of Tunisia.
-
E.
A682
The A682 is a primary road in northern England that runs through Lancashire and North Yorkshire, linking towns such as Rawtenstall and serving as a key route across the Pennines.
- F. None of above.
- G. Unsure - the case is ambiguous/there is not enough information to decide.
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_69d381c16c248190a2fe5b471e584e9c |
completed | April 6, 2026, 9:49 a.m. |
| NER | Named-entity recognition | batch_69d5094cec788190a485c5c9e7cd024a |
completed | April 7, 2026, 1:40 p.m. |
| NED1 | Entity disambiguation (via context triple) | batch_69d8a0094910819094d492c87b31898e |
completed | April 10, 2026, 7 a.m. |
Created at: April 6, 2026, 12:20 p.m.