Triple

T11188514
Position Surface form Disambiguated ID Type / Status
Subject Kohistan District E264735 entity
Predicate capital P234 FINISHED
Object Dassu E264737 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: Dassu | Statement: [Kohistan District, capital, Dassu]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Dassu
Context triple: [Kohistan District, capital, Dassu]
  • A. Dassu chosen
    Dassu is a town in Pakistan’s Khyber Pakhtunkhwa province that serves as the administrative center of Upper Kohistan District in the mountainous Kohistan region.
  • B. Dookudu
    Dookudu is a 2011 Telugu-language action-comedy film widely recognized as one of Mahesh Babu’s biggest commercial hits and a landmark in contemporary mainstream Telugu cinema.
  • C. Sadda
    Sadda is a town in Pakistan’s Khyber Pakhtunkhwa region, known as a local commercial and administrative center in the Kurram District near the Afghan border.
  • D. Cundun
    Cundun is an alternative spelling or variant form of the name Condon, which is used as a surname and place name in various English-speaking regions.
  • E. Dola
    Dola is an island located within Lake Rakshastal in western Tibet, near the sacred Mount Kailash region.
  • 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_69d6aa9eb9248190b20211772621b4bc completed April 8, 2026, 7:21 p.m.
NER Named-entity recognition batch_69d7e8ad143481908d5dacc95837ecfd completed April 9, 2026, 5:58 p.m.
NED1 Entity disambiguation (via context triple) batch_69e483df3c74819090fc6ef26f785539 completed April 19, 2026, 7:27 a.m.
Created at: April 8, 2026, 9:29 p.m.