Triple
T12958569
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Winterfell |
E310078
|
entity |
| Predicate | defensiveStructureType |
P8412
|
FINISHED |
| Object | castle |
—
|
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: castle | Statement: [Winterfell, defensiveStructureType, castle]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: defensiveStructureType Context triple: [Winterfell, defensiveStructureType, castle]
-
A.
defensiveStructure
chosen
Indicates a relationship where one entity functions as a structure built or used to protect, defend, or fortify another entity or area.
-
B.
defensiveSystem
Indicates a system or mechanism whose primary function is to protect, guard, or defend an entity against threats or attacks.
-
C.
defensiveRole
Indicates that an entity serves a protective or guarding function in relation to another entity or context.
-
D.
hasDefensivePositions
Indicates that an entity possesses or controls locations, structures, or arrangements specifically organized for defense against potential threats or attacks.
-
E.
typeOfDefense
Indicates the specific kind or category of defense employed or possessed in a given context.
- 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_69d7bdfb57a88190836b743e2825feca |
completed | April 9, 2026, 2:55 p.m. |
| NER | Named-entity recognition | batch_69d97e59a4c88190907d05b8d57dae89 |
completed | April 10, 2026, 10:48 p.m. |
| PD | Predicate disambiguation | batch_69d97dba57988190b786ffed55687a72 |
completed | April 10, 2026, 10:46 p.m. |
Created at: April 9, 2026, 5:44 p.m.