Harvesting publication data to the institutional repository from Scopus, Web of Science, Dimensions and Unpaywall using a custom R Script

Institutional repositories are established tools for archiving and increasing the visibility and availability of academic outputs. Although the potential benefits of institutional repositories are well researched and many funders and institutions already mandate open access publishing via gold or green open access routes, institutional repositories often struggle with lack of growth and sustained workflows for content recruitment. Institutions have come up with various (and often creative) workflows for populating their repositories, including institutional open access mandates, library-mediated self-archiving, fully or partially automated content harvesting and integrations between repositories and Current Research Information Systems (CRIS). Zayed University launched the ZU Scholars 1 1 https://zuscholars.zu.ac.ae institutional repository in fall 2021. Since the beginning, a semi-automated workflow was introduced to populate the repository with publication data from Scopus, Web of Science, Dimensions and Unpaywall using a custom R script. Full text files are added automatically for all Creative Commons licensed articles. This article describes the data harvesting and conversion process, its current limitations and plans for future development. The article also reviews similar content harvesting projects in the context of institutional repositories