Plasma-MDS, a metadata schema for plasma science with examples from plasma technology

Steffen Franke,Jan Schäfer,Lucian Paulet,Markus M Becker,Deborah O’Connell

doi:10.1038/s41597-020-00771-0

Steffen Franke, Jan Schäfer + Show 3 more

Open Access

https://doi.org/10.1038/s41597-020-00771-0

Copy DOI

Abstract

A metadata schema, named Plasma-MDS, is introduced to support research data management in plasma science. Plasma-MDS is suitable to facilitate the publication of research data following the FAIR principles in domain-specific repositories and with this the reuse of research data for data driven plasma science. In accordance with common features in plasma science and technology, the metadata schema bases on the concept to separately describe the source generating the plasma, the medium in which the plasma is operated in, the target the plasma is acting on, and the diagnostics used for investigation of the process under consideration. These four basic schema elements are supplemented by a schema element with various attributes for description of the resources, i.e. the digital data obtained by the applied diagnostic procedures. The metadata schema is first applied for the annotation of datasets published in INPTDAT—the interdisciplinary data platform for plasma technology.

Highlights

The rapid progress in data science methods for machine-based analysis of big data provides enormous potential for new data driven sciences and the development and optimization of innovative technologies
Metadata represent extra information attached to data that allows people and automated processes to find, access and reuse data
Dublin Core, the DataCite Metadata Schema[23], and DCAT represent fundamental metadata schemata that are widely used for the collection and indexing of general metadata of digital objects, such as title, publication year, and permanent identifier

Summary

Introduction

The rapid progress in data science methods for machine-based analysis of big data provides enormous potential for new data driven sciences and the development and optimization of innovative technologies. In the wide field of plasma science, the application of machine learning methods, e.g. for investigation and control of fusion plasmas, the particle and event identification in high energy physics, and the discovery of space phenomena in astrophysics has been common practice for several years, see[1,2,3] and references therein. First approaches have been published that use machine learning methods for simulation, diagnostics and control of technological plasmas[4,5]. This is of particular interest because technological plasmas are used in many applications and industrial processes. Applications of cold plasmas in medicine include the plasma-based synthesis of biomedical surfaces, wound healing, and cancer treatment[11]

Methods

Results

Conclusion