← Back to catalogue
Research draft

data set

vr.tr.data-set · INF.MED

Let an agent describe datasets and their metadata, help find, evaluate, cite and share datasets, explain licensing and ethics, and support documentation practices.

Thing Registry Information and virtual systems

Research draft, second pass

A second pass drafted this model: the structure a model of this thing needs, and what is known about it in the world. The line under this one says how the second half was obtained - researched against sources, or recalled without web access, in which case nothing here was read anywhere and every claim is a lead to verify. Unreviewed either way.

written by Claude from model knowledge without web access - no source was read, every claim is a lead to verify

Researched by: Claude

Purpose and description

Let an agent describe datasets and their metadata, help find, evaluate, cite and share datasets, explain licensing and ethics, and support documentation practices.

A collection of data organised for use, typically as tables, files or records with a defined structure and documentation, ranging from scientific datasets such as single-cell sequencing data and benchmark datasets to public health datasets, controlled vocabularies and red lists; datasets are described with metadata, licensed, cited and shared through repositories following principles for findability and reuse.

What it is for: Organised data for analysis and reuse.

It can be find and evaluate datasets; document and cite datasets; explain licensing and ethics; support sharing and publishing.

Distinguishing features

Structured collection

Documented

Licensed and citable

Reusable

What it looks like

Not physical; files, tables and repository records.

How it is recognised

Organised collection of data

Documented and licensed

A single measurement or a database system is not a dataset

Related models

is a kind of - category

collection

is a kind of - category

scientific work

is related to - storage

database

is related to - open data parallels

open-source software

In practice

Families and kinds

scientific research datasets

benchmark and training datasets

public and government datasets

reference datasets such as controlled vocabularies and red lists

personal and sensitive datasets

Standards and regulation

FAIR data principles

Data licences such as Creative Commons and ODbL

Data protection laws for personal data

Failure modes and hazards

Undocumented or biased data

Privacy breaches

Licence violations

Also called

Scholia corpus for modelingCOVID-19 datasetcontrolled vocabularyred listbenchmark datasetscRNA-seq datasetlarge datasetspeech datasetsingle cell data setscientific data setdeduplicated data setsoftware citation datasetSurnames present in the PESEL registerdata set with explicit CiTO annotationinstruction datasetopen data setGeoNLP Geographical Name Dictionaryimage datasetgeospatial datasetDatasaurus dozennovel datasetmap tile setpre-shared datasetflow datasetGBIF occurrence datasetIDPhotoSnap Passport Photo SpecificationsFour-sheet map of the boundaries of the Eskola land-allotment community and the lands subject to regulation in the village of Kiikoinen in Suodenniemi and in the village of Eskola in Mouhijärvi, Tyrvää district, Turku and Pori CountyA66:1/1-16 Eskola village; Great Partition 1827–1832data set of a specific countrydataset distributionpoint cloudhierarchical datassnapshotAnscombe's quartetRDF graphRDF datasetKey Sequenced Data Setcross-sectional datacolor channelcard index

+57

Where this came from

wikidata · CC0 1.0

Source repository Registry entry point

Drafted structure

Bundle to layer to finding to question, as the second pass will find it: 4 bundles · 8 layers · 8 findings · 16 questions.

Find Finding datasets.

Discovery.

Search

Locating data.

Search

Discovery.

  1. Which datasets exist for this topic, and where are they hosted? provenance
  2. Which entry fits a specific dataset? action

Evaluate

Assessing quality.

Evaluate

Evaluation.

  1. How can the quality, coverage, bias and provenance of this dataset be assessed? action
  2. Is its documentation adequate? boundary
Use Using datasets.

Rights.

Licence

Licensing.

Licence

Licensing.

  1. Under which licence is this dataset available, and what does it permit? provenance
  2. How should it be cited? action

Ethics

Ethics and privacy.

Ethics

Ethics.

  1. Does this dataset contain personal or sensitive data, and which rules and ethical safeguards apply? boundary
  2. How is consent handled? provenance
Publish Sharing datasets.

Practice.

Document

Documentation.

Document

Documentation.

  1. How should a dataset be documented, with metadata, data dictionaries and provenance, following FAIR principles? action
  2. Which metadata standards apply? provenance

Repository

Repositories.

Repository

Repositories.

  1. Which repositories suit this dataset, and how are identifiers assigned? provenance
  2. How is versioning handled? action
Kinds Special kinds.

Context.

Benchmark

Benchmark and training data.

Benchmark

Benchmarks.

  1. What makes a good benchmark or training dataset, and what are documented problems with existing ones? provenance
  2. How is contamination avoided? provenance

Reference

Reference datasets.

Reference

Reference data.

  1. What are controlled vocabularies and red lists as datasets, and how are they maintained? definition
  2. How can they be used? action

What the second pass must settle

  • Should each major dataset be a separate entry?
  • How should repositories be linked?
  • How should data ethics guidance be linked?
Draft as JSON