Skip to content

This repository highlights the FAIRness quality measures, the common FAIR vocabulary, and the FAIRness Quality Maturity Matrix as defined in Peng et al. (2024).

License

Notifications You must be signed in to change notification settings

gepeng86/FAIRness-QualityMeasures

Repository files navigation

Harmonizing Quality Measures of FAIRness Assessment Towards Machine-Actionable Quality Information

Introduction

This repository highlights the outcomes from Peng et al. (2024), published by the International Journal of Digital Earth of Taylor & Francis for a special collection (Advance in FAIR Geospatial Information Resources).

The FAIRness quality measures established by Peng et al. (2024) are a set of unique core concepts contained in the definitions of the FAIR Principls (Wilkinson et al. 2016), that can be used to evaluate the level of FAIR-compliance, i.e., FAIRness. The core concepts are deduced by decomposing the definitions through a concept mapping approach described in Peng (2023). More information can be found in this folder.

The FAIRness quality measures can serve as common, fundamental pillars of holistic FAIRness assessment workflows. More information can be found in this folder.

A common FAIR vocabulary defined in Peng et al. (2024) is presented with definitions and ontology specifications (working in progress). The common FAIR vocabulary is beneficial for harmonizing FAIRness assessment and reporting towards machine-actionable FAIRness quality information. More information can be found in this folder.

The FAIRness quality maturit matrix (FAIR-QMM) crafted by Peng et al. (2024) is also presented. The FAIR-QMM is a structured, tiered, and progressive approach for evaluating and reporting the degree of FAIR-compliance. It can be used as a FAIRness assessment tool independently and/or as a translator between other FAIRness assessment tools or models. More information can be found in this folder.

Data, Metadata, Infrastructure, and Enterprise Capability are considered as four essential components for FAIR data. The relevant requirements derived from the FAIR Principles are listed in this folder.

Related presentations can be found in this folder.

Release Notes

The contents in this repository is released under the CC-BY 4.0 International license. When using or reusing the described FAIRness quality measures, common FAIR vocabulary, FAIR-QMM, and any other relevant contents, please provide attribution to Peng et al. (2024) and/or this repository.

If you have any improvement suggestions, you can provide feedback by opening an issue or start a discussion in this gitHub repo.

References

Peng, G., G. Berg-Cross, M. Wu, R.R. Downs, S.R. Shrestha, L. Wyborn, N. Ritchey, H.K. Ramapriyan, S.J. Clark, J. Wood, Z. Liu, and A. Marouane. (2024). Harmonizing Quality Measures of FAIRness Assessment Towards Machine-Actionable Quality Information. Int. J. Digit. Earth 17:2390431. https://doi.org/10.1080/17538947.2024.2390431

Peng, G. (2023). Finding harmony in FAIRness. Eos, 23, https://doi.org/10.1029/2023EO230216

Wilkinson, M. D., M. Dumontier, I. J. Aalbersberg, G. Appleton, M. Axton, A. Baak, and others. (2016). The FAIR Guiding Principles for scientific data management and stewardship. Sci. Data, 3, https://doi.org/10.1038/sdata.2016.18

About

This repository highlights the FAIRness quality measures, the common FAIR vocabulary, and the FAIRness Quality Maturity Matrix as defined in Peng et al. (2024).

Topics

Resources

License

Stars

Watchers

Forks

Packages

No packages published