With improved affiliation ingestion and extraction algorithms, the number of research products in the OpenAIRE Graph containing affiliation information has increased to over 79 million (>140Mi in total when counting multiple affiliations). A significant increase, providing the Graph with more data to enrich your monitoring and assessment tasks.

About the Graph: The OpenAIRE Graph is one of the largest and trusted Knowledge Graphs in the scholarly communication domain. Made up of metadata from thousands of trusted sources, it covers an extensive range of research products, including publications, datasets, and software. The Graph also contains links of these products, not only showing relationships between themselves, but also to research organisations, researchers, funders, and other related entities, resulting in an enriched and comprehensive view of the Open Science research landscape.

OpenAIRE Graph Update: Version 6.2.2 Featuring Improved Affiliation Extraction

Affiliations: Links between research organisations and research products (affiliations) are essential for creating added-value services on top of the Graph to help those interested in monitoring the performance of researchers, organisations, or countries in terms of research productivity, scientific impact, and other merits.

The challenge: While affiliation links between organisations and research products are invaluable, they are in reality difficult to identify. Many scholarly communication data sources do not provide this information in their metadata or provide strings in place of persistent identifiers (PIDs) describing the affiliations. 

Persistent identifiers (PIDs) are unique codes essential for linking research entities across databases. In the OpenAIRE Graph, their integration into affiliation links significantly expands the coverage of connected research products.

The solution: To fill this gap, we have worked tirelessly to increase the coverage of included affiliation links via two different methods.

• Mining: We downloaded tens of millions of full-text publications and applied mining algorithms to identify affiliations and match them to records from ROR.org, an open registry of research organisation PIDs.
• Ingestion: We defined aggregation workflows to import affiliation strings from Crossref and PubMed and accurately map them to ROR identifiers. 

What this means for you: This significant development will give users of OpenAIRE MONITOR, Open Science Observatory, and other similar services more comprehensive insights into the research assesmement excersise and the Open Science landscape, facilitating decision-making and helping them better monitor their impact.

What's next: This coverage is only set to increase with every Graph update moving forward, where additional links will be included each time OpenAIRE incorporates a new Crossref and PubMed version, and as we enrich our algorithms. Stay tuned as we continue to enrich and foster Open Science globally, providing you with the tools and data to monitor your impact, evaluate data, increase your visibility, and much more.