{"id":"linked-art/Linked%20Open%20Art%20Data%20Web%20App%20-%20Must-have%20Data%20Sources","relativePath":"linked-art/Linked Open Art Data Web App - Must-have Data Sources.md","title":"Linked Open Art Data Web App (AI) — Must-have Data Sources","markdown":"# Linked Open Art Data Web App (AI) — Must-have Data Sources\n\nPurpose: a curated shortlist of high-leverage open / machine-readable sources to power a **linked open art data** app (entity resolution, enrichment, search, recommendations).\n\n## Recommended pattern (why this stack works)\n- Use a **backbone entity graph** for identity + disambiguation.\n- Use **authority vocabularies** for normalization (names, places, materials, techniques, concepts).\n- Add **institutional collection APIs** for canonical object records, images, provenance, exhibitions.\n- Normalize all ingested data into a common internal model (recommend: **[[Linked Art]]** style modeling).\n\n## Top 10 prioritized sources\n\n### Core art sources\n1. [[Wikidata]]\n   - Best for: global entity backbone (artists, artworks, movements, places, institutions)\n   - Strength: broad linked-data connectivity; identifiers to many external catalogs\n   - Use in product: entity graph + join keys + disambiguation + “sameAs” expansion\n\n2. [[Europeana]]\n   - Best for: aggregation/discovery across European cultural heritage\n   - Strength: Search/Record/Entity APIs for discovery + entity matching\n   - Use in product: broad coverage, cross-institution linking, discovery funnels\n\n3. [[Rijksmuseum]] (Rijksmuseum Data)\n   - Best for: high-quality museum metadata, Europeana-aligned structures\n   - Strength: open museum dataset + APIs/downloads; strong object/creator metadata\n   - Use in product: deep collection depth, high-trust records for training/eval\n\n4. [[The Met Collection API]] (Metropolitan Museum of Art)\n   - Best for: broad Open Access artworks + images\n   - Strength: popular and well-documented; good for rapid prototyping\n   - Use in product: canonical object pages, image-based features, recommendations\n\n5. [[Art Institute of Chicago API]]\n   - Best for: unified public API spanning collection + related content\n   - Strength: consistent API surface; strong for integrating “content around art”\n   - Use in product: object + interpretive content; richer on-site experiences\n\n### High-value linked-data & standards\n6. [[Getty Vocabularies]]\n   - Best for: authority control (names/places/materials/techniques/concepts)\n   - Strength: normalization + disambiguation; improves entity linking quality\n   - Use in product: canonical labels, multilingual expansion, semantic search facets\n\n7. [[Linked Art]] (data model ecosystem)\n   - Best for: interoperability standard for art data publishing/integration\n   - Strength: consistent graph model; reduces bespoke mapping per museum\n   - Use in product: *internal canonical schema* + export format\n\n8. [[Smithsonian American Art Museum LOD]] (SAAM)\n   - Best for: stable URIs for artists/objects; strong American art coverage\n   - Strength: linked open data with resolvable identifiers\n   - Use in product: graph linking, authority-style references, provenance context\n\n9. [[Tate collection data]]\n   - Best for: open collection metadata around artists and artworks\n   - Strength: good coverage; useful for cross-museum linking\n   - Use in product: UK/modern art depth; complementary to Europeana + Wikidata\n\n10. [[American Art Collaborative]] (and similar museum consortia LOD)\n   - Best for: cross-museum linking beyond a single institution\n   - Strength: multi-institution graph context; increases match rates + coverage\n   - Use in product: broader graph + better recommendations across collections\n\n## Practical build order (implementation sequence)\n1) **[[Wikidata]]** for the backbone entity graph + disambiguation.\n2) **[[Getty Vocabularies]]** for controlled terms / normalization.\n3) **[[Europeana]]** for broad aggregation + entity matching.\n4) Add “depth” via: [[Rijksmuseum]], [[The Met Collection API]], [[Art Institute of Chicago API]], [[Smithsonian American Art Museum LOD]], [[Tate collection data]].\n5) Adopt **[[Linked Art]]** as the canonical internal model and map sources into it.\n\n## Notes for AI-powered features (what to extract/standardize)\n- Identity: stable IDs, sameAs links, external identifiers\n- Strings to normalize: person names, place names, titles, materials, techniques\n- Time: production dates (with uncertainty), life dates, period styles\n- Relationships: creator ↔ work, work ↔ movement, work ↔ place, work ↔ institution\n- Assets: image URLs, IIIF manifests (when available), rights statements\n\n## Follow-ups / next steps\n- Create individual notes per source with: endpoints, rate limits, licensing, identifier strategy, mapping notes to [[Linked Art]].\n- Define internal entity types (Artist, Artwork, Place, Movement, Institution, Material, Technique) and linking rules.\n","sections":[{"level":2,"heading":"Recommended pattern (why this stack works)","anchor":"recommended-pattern-why-this-stack-works"},{"level":2,"heading":"Top 10 prioritized sources","anchor":"top-10-prioritized-sources"},{"level":3,"heading":"Core art sources","anchor":"core-art-sources"},{"level":3,"heading":"High-value linked-data & standards","anchor":"high-value-linked-data-standards"},{"level":2,"heading":"Practical build order (implementation sequence)","anchor":"practical-build-order-implementation-sequence"},{"level":2,"heading":"Notes for AI-powered features (what to extract/standardize)","anchor":"notes-for-ai-powered-features-what-to-extract-standardize"},{"level":2,"heading":"Follow-ups / next steps","anchor":"follow-ups-next-steps"}],"html":"<h1 id=\"linked-open-art-data-web-app-ai-must-have-data-sources\">Linked Open Art Data Web App (AI) — Must-have Data Sources</h1>\n<p>Purpose: a curated shortlist of high-leverage open / machine-readable sources to power a <strong>linked open art data</strong> app (entity resolution, enrichment, search, recommendations).</p>\n<h2 id=\"recommended-pattern-why-this-stack-works\">Recommended pattern (why this stack works)</h2>\n<ul><li>Use a <strong>backbone entity graph</strong> for identity + disambiguation.</li><li>Use <strong>authority vocabularies</strong> for normalization (names, places, materials, techniques, concepts).</li><li>Add <strong>institutional collection APIs</strong> for canonical object records, images, provenance, exhibitions.</li><li>Normalize all ingested data into a common internal model (recommend: <strong>[[Linked Art]]</strong> style modeling).</li></ul>\n<h2 id=\"top-10-prioritized-sources\">Top 10 prioritized sources</h2>\n<h3 id=\"core-art-sources\">Core art sources</h3>\n<ol><li>[[Wikidata]]</li></ol>\n<ul><li>Best for: global entity backbone (artists, artworks, movements, places, institutions)</li><li>Strength: broad linked-data connectivity; identifiers to many external catalogs</li><li>Use in product: entity graph + join keys + disambiguation + “sameAs” expansion</li></ul>\n<ol><li>[[Europeana]]</li></ol>\n<ul><li>Best for: aggregation/discovery across European cultural heritage</li><li>Strength: Search/Record/Entity APIs for discovery + entity matching</li><li>Use in product: broad coverage, cross-institution linking, discovery funnels</li></ul>\n<ol><li>[[Rijksmuseum]] (Rijksmuseum Data)</li></ol>\n<ul><li>Best for: high-quality museum metadata, Europeana-aligned structures</li><li>Strength: open museum dataset + APIs/downloads; strong object/creator metadata</li><li>Use in product: deep collection depth, high-trust records for training/eval</li></ul>\n<ol><li>[[The Met Collection API]] (Metropolitan Museum of Art)</li></ol>\n<ul><li>Best for: broad Open Access artworks + images</li><li>Strength: popular and well-documented; good for rapid prototyping</li><li>Use in product: canonical object pages, image-based features, recommendations</li></ul>\n<ol><li>[[Art Institute of Chicago API]]</li></ol>\n<ul><li>Best for: unified public API spanning collection + related content</li><li>Strength: consistent API surface; strong for integrating “content around art”</li><li>Use in product: object + interpretive content; richer on-site experiences</li></ul>\n<h3 id=\"high-value-linked-data-standards\">High-value linked-data &amp; standards</h3>\n<ol><li>[[Getty Vocabularies]]</li></ol>\n<ul><li>Best for: authority control (names/places/materials/techniques/concepts)</li><li>Strength: normalization + disambiguation; improves entity linking quality</li><li>Use in product: canonical labels, multilingual expansion, semantic search facets</li></ul>\n<ol><li>[[Linked Art]] (data model ecosystem)</li></ol>\n<ul><li>Best for: interoperability standard for art data publishing/integration</li><li>Strength: consistent graph model; reduces bespoke mapping per museum</li><li>Use in product: <em>internal canonical schema</em> + export format</li></ul>\n<ol><li>[[Smithsonian American Art Museum LOD]] (SAAM)</li></ol>\n<ul><li>Best for: stable URIs for artists/objects; strong American art coverage</li><li>Strength: linked open data with resolvable identifiers</li><li>Use in product: graph linking, authority-style references, provenance context</li></ul>\n<ol><li>[[Tate collection data]]</li></ol>\n<ul><li>Best for: open collection metadata around artists and artworks</li><li>Strength: good coverage; useful for cross-museum linking</li><li>Use in product: UK/modern art depth; complementary to Europeana + Wikidata</li></ul>\n<ol><li>[[American Art Collaborative]] (and similar museum consortia LOD)</li></ol>\n<ul><li>Best for: cross-museum linking beyond a single institution</li><li>Strength: multi-institution graph context; increases match rates + coverage</li><li>Use in product: broader graph + better recommendations across collections</li></ul>\n<h2 id=\"practical-build-order-implementation-sequence\">Practical build order (implementation sequence)</h2>\n<p>1) <strong>[[Wikidata]]</strong> for the backbone entity graph + disambiguation.</p>\n<p>2) <strong>[[Getty Vocabularies]]</strong> for controlled terms / normalization.</p>\n<p>3) <strong>[[Europeana]]</strong> for broad aggregation + entity matching.</p>\n<p>4) Add “depth” via: [[Rijksmuseum]], [[The Met Collection API]], [[Art Institute of Chicago API]], [[Smithsonian American Art Museum LOD]], [[Tate collection data]].</p>\n<p>5) Adopt <strong>[[Linked Art]]</strong> as the canonical internal model and map sources into it.</p>\n<h2 id=\"notes-for-ai-powered-features-what-to-extract-standardize\">Notes for AI-powered features (what to extract/standardize)</h2>\n<ul><li>Identity: stable IDs, sameAs links, external identifiers</li><li>Strings to normalize: person names, place names, titles, materials, techniques</li><li>Time: production dates (with uncertainty), life dates, period styles</li><li>Relationships: creator ↔ work, work ↔ movement, work ↔ place, work ↔ institution</li><li>Assets: image URLs, IIIF manifests (when available), rights statements</li></ul>\n<h2 id=\"follow-ups-next-steps\">Follow-ups / next steps</h2>\n<ul><li>Create individual notes per source with: endpoints, rate limits, licensing, identifier strategy, mapping notes to [[Linked Art]].</li><li>Define internal entity types (Artist, Artwork, Place, Movement, Institution, Material, Technique) and linking rules.</li></ul>","updatedAt":"2018-10-20T01:46:40.000Z","checksum":"7b7d350fe8a0277522ad6f9b967e40ea414c24b1413f85cc3ca6e0e65cbca592","checksumPrefix":"7b7d350fe8a0","anchorCount":7,"lineCount":82,"rawUrl":"/api/docs/content?path=linked-art%2FLinked%20Open%20Art%20Data%20Web%20App%20-%20Must-have%20Data%20Sources.md","htmlUrl":"/docs?doc=linked-art%2FLinked%20Open%20Art%20Data%20Web%20App%20-%20Must-have%20Data%20Sources.md","apiUrl":"/api/docs/content?path=linked-art%2FLinked%20Open%20Art%20Data%20Web%20App%20-%20Must-have%20Data%20Sources.md"}