Triple

T38657774
Position Surface form Disambiguated ID Type / Status
Subject LIRR Atlantic Branch E939940 entity
Predicate connectsToSubway P158586 FINISHED
Object New York City Subway at Atlantic Terminal 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: New York City Subway at Atlantic Terminal | Statement: [LIRR Atlantic Branch, connectsToSubway, New York City Subway at Atlantic Terminal]
PD Predicate disambiguation gpt-5-mini-2025-08-07
Target predicate: connectsToSubway
Context triple: [LIRR Atlantic Branch, connectsToSubway, New York City Subway at Atlantic Terminal]
  • A. hasSubway
    Indicates that a place is served by or contains a subway (metro) system.
  • B. hasSubwayCode
    Indicates that an entity is associated with a specific subway system code used to identify it within that transit network.
  • C. hasSubwaySymbol
    Indicates that one entity is used as the official subway symbol or icon representing another entity.
  • D. hasSubwayComplex
    Indicates that one entity contains or is associated with a subway complex as part of its structure or facilities.
  • E. hasOverpassOrSubway chosen
    Indicates that one location is connected to or accessible from another via an overpass or a subway passage.
  • F. None of above.

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_69f76ede49648190a48bfe47032a05a3 completed May 3, 2026, 3:50 p.m.
NER Named-entity recognition batch_6a002962f6e081909906d6436bae6407 completed May 10, 2026, 6:44 a.m.
PD Predicate disambiguation batch_6a00284c9c7c8190a77f18a41eee55df completed May 10, 2026, 6:40 a.m.
Created at: May 3, 2026, 4:33 p.m.