Triple
T38462705
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Islamization of Bengal |
E912491
|
entity |
| Predicate | religionReplacedOrSuperseded |
P132710
|
FINISHED |
| Object | Hinduism |
—
|
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: Hinduism | Statement: [Islamization of Bengal, religionReplacedOrSuperseded, Hinduism]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: religionReplacedOrSuperseded Context triple: [Islamization of Bengal, religionReplacedOrSuperseded, Hinduism]
-
A.
successorReligionAtSite
chosen
Indicates that one religion historically replaced or followed another as the primary religion practiced at a specific site.
-
B.
laterRevertedReligion
Indicates that an entity adopted a religion and then subsequently reverted or returned to a previous or different religious affiliation.
-
C.
replacedReligiousRitesOf
Indicates that one set of religious rites or practices has been superseded or taken the place of another.
-
D.
predecessorReligionPolicy
Indicates that a religion policy is the direct predecessor of another religion policy in a temporal or version sequence.
-
E.
dominantReligionLegacy
Indicates that a historically dominant religion has left a lasting influence or heritage on a society, culture, or region.
- 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_69f76e861d8c81908559031dc66e3c15 |
completed | May 3, 2026, 3:49 p.m. |
| NER | Named-entity recognition | batch_6a0061454944819088a0babfe60f69fc |
completed | May 10, 2026, 10:43 a.m. |
| PD | Predicate disambiguation | batch_6a0060b9ee108190b91e8d99a16f2b30 |
completed | May 10, 2026, 10:40 a.m. |
Created at: May 3, 2026, 4:31 p.m.