The purpose

OpenAIRE ELD released two new services: the Data Correction, to enrich repository data by exploiting the vast amount of information made available by OpenAIRE, and the Publication Claim, to ensure that the repository stays up-to-date by automatically discovering new content produced by the institution's researchers in the OpenAIRE Graph, thus reducing the manual input from researchers. Both services are available for repositories running on the latest coming DSpace and DSpace-CRIS versions.

The value of the OpenAIRE ELD

The OpenAIRE services will become much more visible and easier to adopt for the DSpace community. The Notification Broker is immediately actionable in the repository: missing or additional information inferred by OpenAIRE can be reviewed and applied to the local data with one click, e.g. missing or additional metadata such as abstract, subject, project, PIDs, etc. The Publication Claim service allows institutions to grab publications from the OpenAIRE Graph and to match their researchers’ profiles with them. The service also allows the single researcher to improve the system guessing capabilities confirming or rejecting identified matches.

The process

In collaboration with the OpenAIRE technical team, 4Science defined the requirements and the technical details for a public API for the OpenAIRE Broker Service and the data exchange formats. Recommendations about how to improve this service and the existing OpenAIRE REST API has been also collected so that the system will be able to evolve further. Based on these APIs 4Science has designed an extensible solution that allows the direct integration of such services in DSpace and DSpace-CRIS.

The solution offers a dashboard for repository administrators and researchers to manage these services but also a full REST API for further integration. Moreover, the solution adopts a multi-layer architecture abstracting from the specific data provider (OpenAIRE) so that it can be easily extended to other sources if needed. Finally, the intelligence about the actions to perform in response to a data correction and the criteria to be used for scoring the found content can be extended providing an easy path for further improvements of the platform.

The result

Source code

The source code related to the project was merged in the official DSpace-CRIS 7 branch and it will be part of the official release:

• REST Contract: https://github.com/4Science/Rest7Contract/tree/dspace-cris-7
• DSpace REST Backend: https://github.com/4Science/DSpace/tree/dspace-cris-7
• DSpace Angular Frontend: https://github.com/4Science/dspace-angular/tree/dspace-cris-7

DSpace users that want to adopt the solution without endorsing the full DSpace-CRIS extension should look to the following branches:

• REST Contract: https://github.com/4Science/Rest7Contract/tree/dspace-oaire-eld
• DSpace REST Backend: https://github.com/4Science/DSpace/tree/dspace-oaire-eld
• DSpace Angular Frontend: https://github.com/4Science/dspace-angular/tree/dspace-oaire-eld

from where the corresponding PRs RESTContract#156, REST#3143 and Angular#1007 were created to propose the inclusion of these features in the official DSpace codebase.

Related OpenAIRE Products/Services

OpenAIRE Notification Broker: inferred information about the Institution data provided to OpenAIRE are returned back to the Institution so that missing, wrong, and incomplete data can be corrected.

OpenAIRE Research Graph: information provided by other OpenAIRE data providers are analyzed and proposed to the Institutio. When authored by some researchers at the Institution they are suggested for inclusion in the Institutional Repository.

Requirements

A running DSpace (properly patched) or DSpace-CRIS 7+ installation.

The Data Correction Service requires a JSON file produced from a subscription that you need to setup on the OpenAIRE Content Provider dashboard. You can retrieve such file using the command line client to access the Broker enrichments.

See also https://4science.github.io/oaire-eld/#/installation-sourcecode-license.

Team

The 4Science R&D team members that worked on the OpenAIRE ELD project are:

• Andrea Bollini, Lead Developer and Tester - Italy, Rome
• Luigi Andrea Pascarelli, Developer - Italy, Rome
• Giuseppe Digilio, Developer - Italy, Milan
• Matteo Perelli, Developer - Italy, Milan
• Pasquale Cavallo, Developer - Italy, Rome
• Alessandro Martelli, Developer - Italy, Rome
• Pilot at the University of Trieste led by: Jordan Piščanc
• Contact person: Susanna Mornati