Triple
T37633135
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Idukki Hydroelectric Project |
E936409
|
entity |
| Predicate | primaryDam |
P21685
|
FINISHED |
| Object | Idukki Dam |
E2040054
|
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: Idukki Dam | Statement: [Idukki Hydroelectric Project, primaryDam, Idukki Dam]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: primaryDam Context triple: [Idukki Hydroelectric Project, primaryDam, Idukki Dam]
-
A.
primaryFor
Indicates that one entity serves as the main or principal option, resource, or association for another entity among possible alternatives.
-
B.
primaryHeir
Indicates that one entity is the main or principal inheritor of another entity’s estate, title, or rights.
-
C.
primaryComponent
chosen
Indicates that one entity serves as the main or most important component within another entity or system.
-
D.
primarySingle
Indicates that an entity has exactly one main or primary association of the specified type, with no additional concurrent primary associations.
-
E.
primaryDamageType
Indicates the main kind of harm or injury that an action, event, or object is responsible for causing.
- F. None of above.
Provenance (4 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_69f76ed24820819081bafd36e9088701 |
completed | May 3, 2026, 3:50 p.m. |
| NER | Named-entity recognition | batch_6a037c8efcd4819088c2aeead65d93df |
completed | May 12, 2026, 7:16 p.m. |
| NED1 | Entity disambiguation (via context triple) | batch_6a40e0695e5c819092ef5b883f76daee |
completed | June 28, 2026, 8:50 a.m. |
| PD | Predicate disambiguation | batch_6a037a1553e08190bb7424c448cb1f33 |
completed | May 12, 2026, 7:05 p.m. |
Created at: May 3, 2026, 4:18 p.m.