Corpus & data pipeline

Dataset

A dataset is a structured, published collection of records meant to be consumed as data - downloadable and machine-readable.

A dataset is a curated collection of records offered for reuse, usually with a described schema and a license. Unlike a web page meant for reading, a dataset is meant for machines: to be queried, downloaded, or analyzed.

Publishing something as a dataset (with structured metadata) makes it discoverable to data-specific search and to agents.

In GraphCanon

GraphCanon publishes its corpus as an open dataset: a JSON API, a full-corpus text file, and Dataset structured data so tools like Google Dataset Search and AI agents can consume it.

See also

Related terms

Last reviewed 2026-07-09

Command menu

Search tools or jump to a page