Triple
T5038073
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Somali Acacia–Commiphora bushlands and thickets |
E113473
|
entity |
| Predicate | hasFragmentation |
P58854
|
FINISHED |
| Object | increasing habitat fragmentation |
—
|
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: increasing habitat fragmentation | Statement: [Somali Acacia–Commiphora bushlands and thickets, hasFragmentation, increasing habitat fragmentation]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: hasFragmentation Context triple: [Somali Acacia–Commiphora bushlands and thickets, hasFragmentation, increasing habitat fragmentation]
-
A.
hasFragments
Indicates that an entity is composed of, contains, or is associated with one or more smaller constituent parts or pieces.
-
B.
fragmentationLevel
chosen
Indicates the degree to which something is broken into smaller, separate parts or segments.
-
C.
supportsOnlineDefragmentation
Indicates that the subject provides the capability to perform defragmentation operations while remaining online and operational.
-
D.
numberOfFragmentsApprox
Indicates an approximate count of how many fragments or pieces are associated with the subject.
-
E.
hasSplit
Indicates that an entity has undergone a division into two or more distinct parts, groups, or components.
- 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_69bd44384298819089c49e7c330ec7b8 |
completed | March 20, 2026, 12:57 p.m. |
| NER | Named-entity recognition | batch_69bd73daaa788190b670f6c328bfa44f |
completed | March 20, 2026, 4:20 p.m. |
| PD | Predicate disambiguation | batch_69bd71529d608190a53470ba6c14bb1d |
completed | March 20, 2026, 4:09 p.m. |
Created at: March 20, 2026, 1:37 p.m.