How FAIR can you get? Image Retrieval as a Use Case to calculate FAIR Metrics

TL;DR

This paper evaluates how well research data repositories adhere to FAIR principles using image retrieval as a case study, proposing improvements to enhance data FAIRness.

Contribution

It introduces a use-case-centric metric framework for assessing FAIR compliance and provides practical suggestions to improve repository scores.

Findings

No repository currently achieves full FAIR score.

Automatic annotation can improve FAIRness.

Supporting content negotiation enhances data retrieval.

Abstract

A large number of services for research data management strive to adhere to the FAIR guiding principles for scientific data management and stewardship. To evaluate these services and to indicate possible improvements, use-case-centric metrics are needed as an addendum to existing metric frameworks. The retrieval of spatially and temporally annotated images can exemplify such a use case. The prototypical implementation indicates that currently no research data repository achieves the full score. Suggestions on how to increase the score include automatic annotation based on the metadata inside the image file and support for content negotiation to retrieve the images. These and other insights can lead to an improvement of data integration workflows, resulting in a better and more FAIR approach to manage research data.