Triple

T8643446
Position Surface form Disambiguated ID Type / Status
Subject Behdini E204711 entity
Predicate spokenIn P2266 FINISHED
Object Dohuk city E232764 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: Dohuk city | Statement: [Behdini, spokenIn, Dohuk city]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Dohuk city
Context triple: [Behdini, spokenIn, Dohuk city]
  • A. Duhok chosen
    Duhok is a city in the Kurdistan Region of Iraq, known as a growing cultural and economic center surrounded by mountains near the Turkish and Syrian borders.
  • B. Erbil
    Erbil is the capital and largest city of the Kurdistan Region in northern Iraq, known as one of the world’s oldest continuously inhabited urban centers.
  • C. Mosul
    Mosul is a major historic city in northern Iraq, known as a cultural and economic center on the Tigris River.
  • D. Kirkuk
    Kirkuk is a historically significant, oil-rich and ethnically diverse city in northern Iraq that has long been a focal point of political and territorial disputes.
  • E. Diyarbakır
    Diyarbakır is a major historic city in southeastern Turkey, renowned for its extensive black basalt city walls and rich Kurdish and Mesopotamian heritage.
  • 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_69ca834ca1c88190a11ffb0200342fac completed March 30, 2026, 2:06 p.m.
NER Named-entity recognition batch_69cc4798852881909c03c5eadf805e49 completed March 31, 2026, 10:15 p.m.
NED1 Entity disambiguation (via context triple) batch_69cef35ea7108190b164f9c3715be4f5 completed April 2, 2026, 10:53 p.m.
Created at: March 30, 2026, 6:28 p.m.