Modelling research output expressions : metadata schema modelling of publication lifecycles and scholarly entities

Macgregor, George (2023) Modelling research output expressions : metadata schema modelling of publication lifecycles and scholarly entities. In: "Re-Discovery" : Metadata & Discovery Group (MDG) Conference 2023, 2023-09-06 - 2023-09-08, IET Birmingham: Austin Court.

[thumbnail of Macgregor-MDG-2023-modelling-research-output-expressions-metadata-schema-modelling-of-publication]
Preview
 Text. Filename: Macgregor_MDG_2023_modelling_research_output_expressions_metadata_schema_modelling_of_publication.pdf
Accepted Author Manuscript
License: Creative Commons Attribution 4.0 logo
Download (195kB)| Preview
[thumbnail of Macgregor-MDG-2023-Modelling-research-output-expressions-slides]
Preview
 Text. Filename: Macgregor_MDG_2023_Modelling_research_output_expressions_slides.pdf
Final Published Version
License: Creative Commons ShareAlike 4.0 logo
Download (1MB)| Preview
[thumbnail of Macgregor-MDG-2023-marp-slides-Modelling-research-output-expressions] Archive. Filename: Macgregor_MDG_2023_marp_slides_Modelling_research_output_expressions.zip
Final Published Version
License: Creative Commons ShareAlike 4.0 logo
Download (4MB)

Abstract

Originally an OAI-PMH metadata application profile for open repositories in the UK, Rioxx version 2.0 has been widely adopted by repositories in the UK since 2016. Such support has enabled superior discovery potential for repository content owing to Rioxx's evidenced harvesting and aggregation benefits [1]. Emerging from a version 3.0 candidate release, Rioxx: The Research Output Metadata Schema has recently been finalized [2]. Version 3.0 adopts a less UK centric approach to the schema and has introduced significant changes to the way in which research outputs are modelled and described. This includes superior capture of graph relations between other scholarly entities, harnessing greater use of persistent identifiers (PIDs), and reusing semantics from prominent vocabularies while retaining OAI-PMH as the principal data harvesting mechanism [3]. Most notably, the schema borrows aspects of the Functional Requirements for Bibliographic Records (FRBR) conceptual entity–relationship model to understand relationships between scholarly entities, particularly group 1 entities pertaining to works and expressions [4]. This, combined with better encoding of these relational associations within metadata, means Rioxx can make a useful contribution to the evolution of open scholarly graphs, including the burgeoning 'PID graph'. This paper explores the modelling of research output expressions within repository metadata schema. Starting with consideration of the Scholarly Works Application Profile (SWAP) [5] and using the recently finalized Rioxx v3.0 schema as case studies, we consider the need for, and the importance of, superior modelling of scholarly works, particularly within open repositories, and consider some of the social-technical impediments to delivering schema which are more adaptable, work agnostic, and PID-centric. We demonstrate how such repository metadata can greatly enrich the formal PID graph thereby potentially unlocking new areas of scientific study; but also enhance user discovery of related research entities, most notably of research publication expressions, datasets, software, projects, and grants. The paper will also explore areas where FRBR thinking needs reactivating within repository contexts. Despite its general acceptance elsewhere, even with its limitations understood [6], experience within open repositories suggests that a commensurate conceptual shift remains elusive in metadata thinking and practice.

Persistent Identifier

https://doi.org/10.17868/strath.00085166
CORE (COnnecting REpositories)