Triple

T15496878
Position Surface form Disambiguated ID Type / Status
Subject Odzi River E378841 entity
Predicate mouthOf P1008 FINISHED
Object Save River E324747 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: Save River | Statement: [Odzi River, mouthOf, Save River]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07 
Target entity: Save River
Context triple: [Odzi River, mouthOf, Save River]
  • A. Save River chosen
    The Save River is a major waterway in southeastern Africa that flows through Zimbabwe and Mozambique before emptying into the Indian Ocean.
  • B. Save River
    The Save River is a watercourse in southwestern France that flows through the Gers department before joining the Garonne River.
  • C. Jump River
    Jump River is a tributary waterway in Wisconsin that feeds into the larger Chippewa River system.
  • D. Every River
    "Every River" is a song by the Scottish Celtic rock band Runrig, known for its emotive melody and themes of longing and connection.
  • E. Keep River
    The Keep River is a river in the Northern Territory of Australia that flows through remote, sparsely populated landscapes before reaching the Joseph Bonaparte Gulf.
  • F. None of above.
  • G. Unsure - the case is ambiguous/there is not enough information to decide.

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_69d85cd53a7c819080f5b9042c4c199e completed April 10, 2026, 2:13 a.m.
NER Named-entity recognition batch_69e03faecd60819091eeaa56c9c8f67d completed April 16, 2026, 1:47 a.m.
NED1 Entity disambiguation (via context triple) batch_69ff3665769c8190be1af51a82a5e75f completed May 9, 2026, 1:28 p.m.
Created at: April 10, 2026, 3:52 a.m.