Triple

T6237305
Position Surface form Disambiguated ID Type / Status
Subject South African national rail network E139507 entity
Predicate abbreviation P43 FINISHED
Object TFR
TFR is the state-owned freight rail division of Transnet that operates and manages South Africa’s national rail freight network.
 E577526 NE FINISHED

How this triple was built (4 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: TFR | Statement: [South African national rail network, abbreviation, TFR]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07 
Target entity: TFR
Context triple: [South African national rail network, abbreviation, TFR]
  • A. TFFR
    TFFR is the ICAO airport code for Pointe-à-Pitre International Airport, the main international gateway to Guadeloupe in the Caribbean.
  • B. TR
    TR is the common abbreviation for the Textus Receptus, a historically influential printed Greek New Testament text that underlies many early Protestant Bible translations.
  • C. TR
    TR is the two-letter ISO 3166-1 alpha-2 country code assigned to Turkey for international standardization and referencing.
  • D. TR
    TR is the postcode area covering Truro and much of Cornwall in southwest England.
  • E. TFI
    TFI is the World Health Organization’s Tobacco Free Initiative, a program dedicated to reducing global tobacco use and its health impacts.
  • F. None of above. chosen
  • G. Unsure - the case is ambiguous/there is not enough information to decide.
NEDg Description generation gpt-5.1
Instruction
Generate a one-sentence description of the target entity. 
You are given a context triple in the form (subject, predicate, object), where the object is the target entity. 
# Instructions
Use the triple to infer relevant information about the entity. Describe the entity based on what is most defining, well-known. 
Avoid repeating the information from the triple, unless really essential.
# Response Format
Return only the sentence: "Description: [one-sentence description of the target entity]"
Input
Entity: TFR
Triple: [South African national rail network, abbreviation, TFR]
Generated description
TFR is the state-owned freight rail division of Transnet that operates and manages South Africa’s national rail freight network.
NED2 Entity disambiguation (via description) gpt-5-mini-2025-08-07 
Target entity: TFR
Target entity description: TFR is the state-owned freight rail division of Transnet that operates and manages South Africa’s national rail freight network.
  • A. TFFR
    TFFR is the ICAO airport code for Pointe-à-Pitre International Airport, the main international gateway to Guadeloupe in the Caribbean.
  • B. TR
    TR is the common abbreviation for the Textus Receptus, a historically influential printed Greek New Testament text that underlies many early Protestant Bible translations.
  • C. TR
    TR is the two-letter ISO 3166-1 alpha-2 country code assigned to Turkey for international standardization and referencing.
  • D. TR
    TR is the postcode area covering Truro and much of Cornwall in southwest England.
  • E. TFI
    TFI is the World Health Organization’s Tobacco Free Initiative, a program dedicated to reducing global tobacco use and its health impacts.
  • F. None of above. chosen

Provenance (5 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_69c008b0e7ac8190808a59573ee646f3 completed March 22, 2026, 3:20 p.m.
NER Named-entity recognition batch_69c063021258819093a9237041816638 completed March 22, 2026, 9:45 p.m.
NED1 Entity disambiguation (via context triple) batch_69c20dfbf42c8190842a471db4ff3de0 completed March 24, 2026, 4:07 a.m.
NEDg Description generation batch_69c215efd48c81908365f0525cb6e3dc completed March 24, 2026, 4:41 a.m.
NED2 Entity disambiguation (via description) batch_69c21654dfac8190a5e985d539e2bcb4 completed March 24, 2026, 4:43 a.m.
Created at: March 22, 2026, 4:23 p.m.