AI-Driven Search System for Nationwide Materials Data Platform

1.1 Data Processing Infrastructure (Cloud, Large-Scale Data Processing Systems, Machine Learning Frameworks, Databases, Data Structure)
1.5 Data Sharing, Data Co-Creation (Search, Data Linkage, Data Authoring. Meta-data Creating)
2.1 Material (Material Informatics, Remote Experiment, Laboratory Automation, etc.)

Masatoshi Hanai

Information Technology Center

Project Assistant Professor

In recent years, the importance of data in materials science has been increasingly emphasized. There are two major types of materials data: one is simulation data generated by theoretical calculations using supercomputers, and the other is experimental data obtained from actual materials testing equipment. In this project, we aim to centrally aggregate materials data—traditionally managed separately in theoretical and experimental domains—into the mdx platform. Furthermore, we will build an AI-based search infrastructure to effectively utilize this integrated and heterogeneous dataset, and implement it in a practical nationwide operation.

Research collaborators

The University of Tokyo
The Japan Atomic Energy Agency
Hiroshima University

Related publications

M. Hanai, R. Ishikawa, M. Kawamura, M. Ohnishi, N. Takenaka, K. Nakamura, D. Matsumura, S. Fujikawa, H. Sakamoto,
Y. Ochiai, T. Okane, S. Kuroki, A. Yamada, T. Suzumura, J. Shiomi, K. Taura, Y. Mita, N. Shibata, Y. Ikuhara
“ARIM-mdx Data System: Towards a Nationwide Data Platform for Materials Science” in IEEE BigData 2024