Triple
T1065046
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Hurricane Katrina |
E22990
|
entity |
| Predicate | thirdLandfallNear |
P20108
|
FINISHED |
| Object | near the Louisiana–Mississippi border |
—
|
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: near the Louisiana–Mississippi border | Statement: [Hurricane Katrina, thirdLandfallNear, near the Louisiana–Mississippi border]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: thirdLandfallNear Context triple: [Hurricane Katrina, thirdLandfallNear, near the Louisiana–Mississippi border]
-
A.
landfallLocation
Indicates the geographic location where a storm or similar weather system first makes landfall.
-
B.
landedNear
chosen
Indicates that one entity has come to rest on a surface or location in close proximity to another specified entity or reference point.
-
C.
nearestSea
Indicates that one location is the closest sea to a given place compared to all other seas.
-
D.
hasNearbyCoast
Indicates that one location is situated close to a coastline or seashore.
-
E.
hasIslandNearby
Indicates that one location is situated close to an island in geographic space.
- 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_69a493dada0481909c43649f9843ea91 |
completed | March 1, 2026, 7:30 p.m. |
| NER | Named-entity recognition | batch_69a4b90f91248190ace1534a51b82bdd |
completed | March 1, 2026, 10:09 p.m. |
| PD | Predicate disambiguation | batch_69a4b7359eb881909c868a558861cc18 |
completed | March 1, 2026, 10:01 p.m. |
Created at: March 1, 2026, 7:42 p.m.