Triple

T9899056
Position Surface form Disambiguated ID Type / Status
Subject Azure Cosmos DB E182241 entity
Predicate supportsAPI P203 FINISHED
Object Table API
Table API is a schema-less, key-value data access interface in Azure Cosmos DB designed for scalable, low-latency storage and retrieval of tabular data.
E828308 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: Table API | Statement: [Azure Cosmos DB, supportsAPI, Table API]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: Table API
Context triple: [Azure Cosmos DB, supportsAPI, Table API]
  • A. Table API
    Table API is Apache Flink’s high-level, declarative interface for expressing data processing and analytics using relational-style operations on streaming and batch data.
  • B. Tabularium
    The Tabularium was the official records office of ancient Rome, a monumental state archive building overlooking the Roman Forum.
  • C. SQL API
    SQL API is a query interface in Apache Flink that lets users define streaming and batch data processing logic using standard SQL syntax.
  • D. Tabular Editor
    Tabular Editor is a specialized development tool for creating, managing, and optimizing tabular models used in platforms like Azure Analysis Services and Power BI.
  • E. DataSet API
    DataSet API is Apache Flink’s now-legacy batch processing API for defining and executing scalable, distributed data transformations.
  • 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: Table API
Triple: [Azure Cosmos DB, supportsAPI, Table API]
Generated description
Table API is a schema-less, key-value data access interface in Azure Cosmos DB designed for scalable, low-latency storage and retrieval of tabular data.
NED2 Entity disambiguation (via description) gpt-5-mini-2025-08-07
Target entity: Table API
Target entity description: Table API is a schema-less, key-value data access interface in Azure Cosmos DB designed for scalable, low-latency storage and retrieval of tabular data.
  • A. Table API
    Table API is Apache Flink’s high-level, declarative interface for expressing data processing and analytics using relational-style operations on streaming and batch data.
  • B. Tabularium
    The Tabularium was the official records office of ancient Rome, a monumental state archive building overlooking the Roman Forum.
  • C. SQL API
    SQL API is a query interface in Apache Flink that lets users define streaming and batch data processing logic using standard SQL syntax.
  • D. Tabular Editor
    Tabular Editor is a specialized development tool for creating, managing, and optimizing tabular models used in platforms like Azure Analysis Services and Power BI.
  • E. DataSet API
    DataSet API is Apache Flink’s now-legacy batch processing API for defining and executing scalable, distributed data transformations.
  • 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_69ca82876f8081909cf75df0f99bb13f completed March 30, 2026, 2:02 p.m.
NER Named-entity recognition batch_69cdb4adc03481909e0f657db01e5bab completed April 2, 2026, 12:13 a.m.
NED1 Entity disambiguation (via context triple) batch_69d1eb1b9534819093c5150f1ed8f685 completed April 5, 2026, 4:54 a.m.
NEDg Description generation batch_69d1ecb3e4a08190add9b971d96f331d completed April 5, 2026, 5:01 a.m.
NED2 Entity disambiguation (via description) batch_69d1ed413edc81908177be16d5127a58 completed April 5, 2026, 5:04 a.m.
Created at: March 30, 2026, 8:40 p.m.