Presentation Information
[U04-P03]AMIDER: A Multidisciplinary Research Data Catalog and Its Initiatives for Data Discovery and Curation★Invited Papers
*Masayoshi Kozai1, Yoshimasa Tanaka1, Shuji Abe2, Yasuyuki Minamiyama3, Atsuki Shinbori7, Qi Zhang1, Akira Kadokura1, Toshiki Shimbaru4, Naoto Kai5, Hayato Tomisu6, Tomoki Yoshihisa6 (1.Joint Support-Center for Data Science Research, Research Organization of Information and Systems, 2.International Research Center for Space and Planetary Environmental Science, Kyushu University, 3.Institute of Social Science, The University of Tokyo, 4.Faculty of Medical Science, Fukuoka International University of Health and Welfare, 5.D3 Center, The University of Osaka, 6.Graduate School of Data Science, Shiga University, 7.Institute for Space-Earth Environment Research, Nagoya University)
Keywords:
Research data management,Open data,Data publication,Metadata
Sharing research data, such as scientific experiment data, forms a basis for studying complex systems in nature across Earth and planetary sciences. The AMIDER (Advanced Multidisciplinary Integrated-database for Exploring new Research) website was released at https://amider.rois.ac.jp in April 2024 to demonstrate an advanced data-sharing platform (Kozai et al., 2025, Data Science Journal). It targets non-expertized users interested in connecting to diverse scientific fields. Over 15,000 metadata records, mainly in polar science, including solar-terrestrial observations, meteorological data, meteorite samples, and animal specimens, have been registered through a collaboration with polar science communities. AMIDER is characterized as an integrated system applicable to different data types, such as physical observation data and specimens, and a user-friendly web application. Users can quickly grasp diverse research data at a glance through AMIDER's catalog view, inspired by web marketing and featuring thumbnail images and snippets. Multidisciplinary data curation and management are also AMIDER’s uniqueness, where multiple metadata schemas (the SPASE and ISO GMD data models) are integrated. For the content data, the AMIDER system handles standardized, self-describing data formats (CDF and NetCDF), enabling useful functions such as format conversion. Data curation software, such as metadata handling tools, is developed and released (https://github.com/AMIDER-dev). More advanced attempts, such as applying text mining techniques to metadata, are also ongoing. These approaches provide new data promotion techniques, including visualisation of cross-data knowledge graphs, and offer insights for improving metadata quality. In this presentation, we introduce the concept, design, operation status, and experimental efforts of the AMIDER project.