Starrydata2 Manual
A guide to searching, browsing, curating, and downloading data in Starrydata2.
Official site: https://starrydata.nims.go.jp/starrydata2/
Introduction
Starrydata2 is a web system that extracts experimental data from figures in published papers on inorganic materials (primarily thermoelectric, magnetic, and other physical properties), converts them into structured curves, and stores them alongside sample information such as composition and fabrication process. As of 2026, the database contains more than 194,000 curves, 82,000 samples, and 13,000 papers (see the homepage statistics for the latest numbers).
All data are available free of charge for both commercial and non-commercial use, and can be downloaded as CSV or JSON files from the web. Any registered researcher can participate in data registration (curation).
Manual chapters
Sign up, log in, choosing a username, and resetting your password.
Paper search, composition search, filters, sample detail pages, and how to read plot views.
Importing papers, extracting curves with StarryDigitizer, entering sample information, and editing guidelines.
Downloading CSV / JSON, Python code examples for loading data, and how to cite.
Who this manual is for
- External researchers and students: New users of Starrydata2. We recommend reading chapters in order: Account Registration → Search → Download.
- Data curators: Researchers who want to register data from their own papers or their group's papers. Focus on Chapter 3.
- API / programmatic users: Researchers who want to handle bulk data for machine learning. Focus on Chapter 4.
Related resources
- Starrydata: from published plots to shared materials data (STAM:M, 2025) — Peer-reviewed paper describing the entire system.
- Starrydata 3D Thermoelectric Materials Explorer — 3D visualization tool for thermoelectric performance.
- NIMS MDR — Starrydata — Permanent URL archive of the project paper.