Metadata for cataloguing datasets from materials science and related domains

There are vast amounts of data available within materials science and only a fraction of these are published. Such data can come from research and development or manufacturing and span experimental data, simulations and may be from diverse laboratories. Indeed, much data is often lost or inaccessible because the metadata is often lacking and inconsistent, and experimental protocols can vary.

[We have written about the importance data catalogues in materials science previously over on our Semantic Materials site]

To help address these concerns, a considerable body of work has been undertaken by the RDA Working Group “Harmonised terminologies and schemas for FAIR data in materials science and related domains” supported by RDA TIGER and RDA Europe. This work has identified a set of strategies to improve the FAIR maturity of materials data, for example, regarding the need to have “Rich metadata… to allow discovery”.

A primary outcome from this work has been a report that, with a global perspective, reviews the state of the art, with a particular focus on the current two most used frameworks, DCAT and Schema.org, and the use of metadata for materials data cataloguing in general.

The report looks at how scientists organise and describe materials science datasets so they can be easily found, shared, and reused. It provides an overview of international efforts in the US, Europe, Japan, and Korea to create common standards for metadata and outlines some implementation cases. The report also highlights successes, gaps, and recommendations for harmonising metadata and how this could be achieved across countries and disciplines in the future.