The volume of data is growing exponentially every year, making efficient storage, retrieval, and reuse increasingly critical in today’s technological landscape. This is true also in Materials Science, where the huge number of experimental and computational techniques to study, characterize, and predict properties of materials results in a large variety of datasets and representations.

The FAIR principles [1] provide a solution to this challenge. Adhering to these principles helps create reliable systems that contribute to sustainable development. According to FAIR guidelines, data should be described by rich metadata (F2) to ease findability and enhance data sharing possibilities [2]. Moreover, to favour data findability, interoperability, and reuse, metadata should use vocabularies that comply with FAIR principles (I2) [2].

The quality and quantity of metadata plays a crucial role in the data exchange and reusability. As an example, detailed descriptions of data generated using Electron Microscopy techniques may include hundreds of terms in the metadata schema. The maintenance of this huge amount of terms and controlled vocabularies is definitely a challenging and time-consuming task.

Controlled vocabularies define terms within a specific domain, including hierarchical structures, descriptions, synonyms, and related concepts. To describe the terminology and semantics of Electron Microscopy techniques at the application level, the Helmholtz Metadata Collaboration (HMC) [3] developed the Electron Microscopy glossary (EM glossary) [4] using the Web Ontology Language (OWL). Each term of the EM glossary has an Internationalized Resource Identifier (IRI) associated to it, which allows it to be unambiguously resolved in a machine-readable way. In order to enhance the FAIRness of metadata describing Electron Microscopy resources, these IRIs can be integrated into metadata schemas.

The Nanoscience Foundries and Fine Analysis (NFFA)-Europe Pilot (NEP) [5], the Helmholtz Joint Lab “Integrated Model and Data-driven Materials Characterization” (JL-MDMC) [6], and the National Research Data Infrastructure for Materials Science & Engineering (NFDI-MatWerk) [7] collaboratively created the Scanning Electron Microscopy (SEM) metadata schema, to describe SEM images in order to facilitate data reuse and exchange [8].

Our goal is to integrate the EM glossary terminology into the SEM metadata schema using the IRIs to enhance the centralized maintenance of the terms and the machine readability and actionability of the descriptive metadata. An application case which demonstrates the advantages of this approach is the metadata enrichment process using a metadata editor: the form automatically resolves the terms in the metadata schema and retrieves their properties (e.g., name, description, controlled list of items) from the EM glossary, supporting the user in filling the corresponding fields.

The integration of community-established vocabularies in metadata schemas improves the user experience at creation time, helps automatically propagate the changes in the metadata schema terms, simplifies future data discovery and reuse across systems, facilitating interoperability and integration within complex ecosystems.

References:

[1] M. D., Wilkinson, et al. Scientific Data, 2016, 3, 160018

[2]https://www.go-fair.org/fair-principles/

[3] https://helmholtz-metadaten.de/de

[4] https://emglossary.helmholtz-metadaten.de

[5] https://nffa.eu/

[6] https://jl-mdmc-helmholtz.de/

[7] https://nfdi-matwerk.de

[8] R. Joseph, et al., CEUR Workshop Proceedings, 2021, 3036, 265-277