FAIR-DI European Conference on Data Intelligence 2024

FAIR Data Quality Metrics in NOMAD

Oct 28, 2024, 8:00 PM

1h

Karoline, ground floor (Achat Hotel)

Poster Developing strategies for scale-bridging workflows in computational materials science Poster Session

Speaker

Nathan Daelman (Institut für Physik, Humboldt-Universität zu Berlin, Berlin)

Description

The FAIR principles (Findable, Accessible, Interoperable, Reusable) serve as a reference for assessing the quality of data storage and publication [1]. NOMAD [nomad-lab.eu] [2, 3] is an open-source data infrastructure for materials science data that is built upon these principles. In this contribution, we will demonstrate the interplay between high-quality data and knowledge using the functionalities provided by NOMAD and with DFT as an example case. In particular, we will showcase the dynamic and flexible metadata framework, designed for a clearer, more customizable navigation of the zoo of density functionals. We will then show how precision and accuracy metrics are represented within this framework, and how they can be linked to benchmark datasets. Finally, we will present a brief outlook on the future of NOMAD as a platform that fosters an interconnected research community and engaged scientific discourse.
[1] Wilkinson, M. D. et al., Sci. Data 3, 160018 (2016).
[2] Scheffler, M. et al., Nature 604, 635-642 (2022).
[3] Scheidgen, M. et al., JOSS 8, 5388 (2023).