Triple
T2600313
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Tantallon Castle |
E58325
|
entity |
| Predicate | primaryDefence |
P35255
|
FINISHED |
| Object | massive curtain wall |
—
|
LITERAL 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: massive curtain wall | Statement: [Tantallon Castle, primaryDefence, massive curtain wall]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: primaryDefence Context triple: [Tantallon Castle, primaryDefence, massive curtain wall]
-
A.
defends
Indicates that one entity protects or supports another entity against attack, criticism, or harm.
-
B.
defense
chosen
Indicates an action or relationship in which an entity protects, guards, or resists against a threat, attack, or criticism from another entity.
-
C.
defenseResult
Indicates the outcome or consequence of a defensive action or strategy in response to an attack or threat.
-
D.
defender
Indicates a relationship where one entity protects, guards, or supports another entity against threats, attacks, or criticism.
-
E.
defenceHandledBy
Indicates that responsibility for managing or conducting a defence is assigned to or carried out by a particular agent or entity.
- 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_69ab4ac14040819098b13f4a27d5c8ff |
completed | March 6, 2026, 9:44 p.m. |
| NER | Named-entity recognition | batch_69abd457564c819080d8c8818c02545a |
completed | March 7, 2026, 7:31 a.m. |
| PD | Predicate disambiguation | batch_69abd0d4e8648190b612eb09aa085451 |
completed | March 7, 2026, 7:16 a.m. |
Created at: March 6, 2026, 9:49 p.m.