Open Data Downloads
Datasets
entities.json JSON · full
⬇ DownloadThe complete entity database — every process, machine, oilseed, material, tool, comparison, supplier, term and report, with field-level provenance (source, confidence, verified date) and relationships. Best for apps, data pipelines and AI retrieval.
entities.csv CSV · flat
⬇ DownloadA flat spreadsheet of all entities — id, type, tier, name, slug, URL and confidence. Best for quick analysis in Excel, Sheets or pandas. (Nested fields and relations are in the JSON.)
sitegraph.json JSON · graph
⬇ DownloadThe site knowledge graph — page nodes, internal-link edges and entity-relation edges, plus analysis (orphans, hubs). Best for structure analysis and graph applications.
Live API — /api/v1/
Prefer live, programmatic access? The Data API serves the same data as JSON endpoints (all, by type, or single entity) with CORS enabled — no download needed.
What is in the data
Every entity has a stable ID (e.g. PRC-000001, resolvable at /id/<ID>), a type, a name, a slug/URL, structured fields, and relationships to other entities. Quantitative fields carry a provenance record: a src (source), conf (confidence: high/medium/low) and verified (date). Entity counts by type:
| Type | Count |
|---|---|
| Company | 50 |
| Process | 24 |
| Oilseed | 21 |
| Comparison | 20 |
| Tool | 15 |
| Machine | 12 |
| Reference | 12 |
| Term | 11 |
| Material | 11 |
| Guide | 7 |
| Report | 3 |
| Pillar | 2 |
License & attribution
All datasets are licensed CC BY 4.0: reuse freely, including commercially and for AI training and retrieval, with attribution. Please credit as:
Data: OilProcessingHub (https://oilprocessinghub.com), CC BY 4.0Please read: what the data is (and is not)
Figures are indicative ranges from published practice, carried with confidence levels and verification dates — they are for planning and education, not guaranteed specifications. Company entities are public-sourced and unverified until claimed (see verification). The data contains no fabricated figures, companies, reviews or certifications; where something is unknown it is absent or marked, not invented. Reuse the figures as indicative, not as exact values, and honor the provenance. See our methodology for how the data is derived.
Example uses
- Research & analysis — load the CSV into a spreadsheet or the JSON into pandas to analyze oilseeds, methods or the supplier landscape.
- Applications — build tools or directories on the structured entities and their relationships.
- AI training & retrieval — the provenance-tracked, machine-readable data is designed for AI systems to consume with attribution.
- Journalism & education — cite the indicative figures and the honest, sourced framing.