At the recent International Digital Curation Conference (IDCC25), OpenAIRE presented a lightning talk on "Combining Algorithms and Human Expertise: OpenAIRE's Entity Disambiguation Method." Delivered by Stefania Amodeo, the presentation highlighted OpenAIRE's innovative approach to maintaining high-quality metadata in the OpenAIRE Graph.
The Importance of Entity Disambiguation in Open Science
In today's complex research landscape, one of the biggest challenges for open scholarly communication is the accurate identification of research entities. With multiple providers contributing metadata about authors, organizations, data sources, and research outputs, maintaining consistency becomes crucial. To address this, OpenAIRE employs a hybrid approach that combines automated deduplication algorithms with human curation expertise to resolve ambiguities in metadata. This method ensures that research entities—such as research products, organisations, and data sources—are properly identified and connected, improving the reliability and discoverability of scholarly information.
The talk also introduced OpenOrgs, a specialized OpenAIRE service that harnesses metadata curator expertise to refine organizational affiliations within the scholarly knowledge graph. By addressing inconsistencies and errors in institutional data, OpenOrgs enhances the precision of research impact assessments and analytics.
IDCC25: A Forum for Digital Curation Innovation
IDCC25, marking twenty years since its first edition, provided a platform to reflect on the progress and emerging challenges in digital curation and research data management. The event facilitated discussions on FAIR (Findable, Accessible, Interoperable, and Reusable) research practices, sustainability in data curation, and the impact of legal and ethical considerations in AI-driven metadata management.
Key Takeaways for Open Science
- Collaboration opportunities: While many organizations have established strong individual practices, there is significant potential for enhanced digital preservation through cross-institutional partnerships.
- Political uncertainty: The conference highlighted how shifting political landscapes pose challenges to maintaining long-term research infrastructure investments.
- The future of curation: Discussions emphasized the evolving skills and training required for metadata curators and data managers in an increasingly AI-driven environment.
As a leading infrastructure for Open Science, OpenAIRE plays a crucial role in digital curation and metadata quality. IDCC25 reinforced the significance of OpenAIRE’s contributions, particularly in ensuring that research outputs are accurately linked and preserved for long-term use. By refining entity disambiguation techniques, OpenAIRE strengthens the global research ecosystem, making scholarly communication more transparent and efficient.
Explore more:
