Since the launch of the Materials Genome Initiative in the USA in 2011, several open access databases have been developed covering specific classes and properties of materials. In parallel to this activity, many effective applications of data-driven methods have been applied in materials science [1-3]. However, none of these existing databases has built-in machine learning (ML) and artificial intelligence (AI) functionality that is accessible and easy to use for end-users without a data science background. Most of these ML and AI applications are only available in the form of separate Python/R scripts on local PCs and GitHub repositories and require a deep understanding of these methods to use them. Therefore, there is a need to integrate existing materials databases and developed AI methods into a so-called Integrated Intelligent Information System (IIIS) for effective data-driven modelling and discovery of new materials.

In this work, the Chair of Materials Discovery and Interfaces at the Ruhr-University Bochum has proposed a solution for an IIIS, which is currently under active development in our group. The proposed IIIS integrates existing experiments and simulations from heterogeneous data sources with additional built-in intelligent ML and AI methods and tools. Such an implementation of an information system in the field of materials science and engineering provides a comprehensive solution for fast and robust data-driven modelling and discovery of new materials. The system is based on open-source standards such as OpenID Connect, OpenAPI and FAIR principles, and can be easily extended by introducing new materials data types, experiments, and simulations, and by connecting new ML and AI algorithms as a service to process materials data [4, 5].

References
[1] U. Nwachukwu, et al., MSMSE, 2022, 30, 025009,
[2] M. Ahmed, et al., MSMSE, 2021, 29, 055012,
[3] B. Bocklund, et al., MRS Communications, 2019, 9, 618-627,
[4] L. Banko, A.Ludwig, ACS Combinatorial Science, 2020, 8 (22), 401–409,
[5] Senko O.V., et al., Lobachevskii Journal of Mathematics, 2023, 44, 188-197.