Triple

T16418136
Position Surface form Disambiguated ID Type / Status
Subject Mara Region E398740 entity
Predicate contains P35 FINISHED
Object Mara River 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: Mara River | Statement: [Mara Region, contains, Mara River]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Mara River
Context triple: [Mara Region, contains, Mara River]
  • A. Mara River chosen
    The Mara River is a major waterway in East Africa that flows through Kenya and Tanzania, supporting rich wildlife ecosystems and the famed wildebeest migration.
  • B. Mara River
    The Mara River is a watercourse flowing through the Maramureș region of northern Romania, contributing to the area's distinctive mountainous and rural landscape.
  • C. Bua River
    The Bua River is a major river in Malawi that drains part of the country’s central region before emptying into Lake Malawi.
  • D. Sabi River
    The Sabi River is a major river in southeastern Africa that flows through Zimbabwe and Mozambique before emptying into the Indian Ocean.
  • E. Magaguadavic River
    The Magaguadavic River is a river in southwestern New Brunswick, Canada, that flows southward from inland lakes to the Bay of Fundy.
  • F. None of above.
  • G. Unsure - the case is ambiguous/there is not enough information to decide.

Provenance (2 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_69d87f2b9024819085c20e52de95d583 completed April 10, 2026, 4:40 a.m.
NER Named-entity recognition batch_69e328798a488190a5fad01c3c95584c completed April 18, 2026, 6:45 a.m.
Created at: April 10, 2026, 5:09 a.m.