Triple

T8738899
Position Surface form Disambiguated ID Type / Status
Subject K. Balachander E207453 entity
Predicate employer P7 FINISHED
Object Doordarshan E128419 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: Doordarshan | Statement: [K. Balachander, employer, Doordarshan]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Doordarshan
Context triple: [K. Balachander, employer, Doordarshan]
  • A. Doordarshan chosen
    Doordarshan is India’s public service broadcaster and one of the country’s oldest and most widely accessible television networks.
  • B. Prasar Bharati
    Prasar Bharati is India’s autonomous public service broadcasting corporation, overseeing major national radio and television networks.
  • C. Sansad TV
    Sansad TV is an Indian parliamentary television channel that provides live coverage and analysis of proceedings from the country's legislative bodies along with related public affairs programming.
  • D. Vijay TV
    Vijay TV is a popular Tamil-language television channel in India known for its entertainment shows, reality programs, and serials.
  • E. Pakistan Television Corporation
    Pakistan Television Corporation is Pakistan’s state-owned public television broadcaster, operating multiple national and regional channels that air news, entertainment, sports, and cultural programming.
  • 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_69ca835a03a081909d4d4cd01a18c9fb completed March 30, 2026, 2:06 p.m.
NER Named-entity recognition batch_69cc5d470c8c81909ead395ef704c6ba completed March 31, 2026, 11:48 p.m.
NED1 Entity disambiguation (via context triple) batch_69cf42d5dd508190854fbbc2541aa819 completed April 3, 2026, 4:32 a.m.
Created at: March 30, 2026, 6:38 p.m.