GESIS Knowledge Graph: Connecting a domain-specific KG to the NFDI and beyond

The GESIS Knowledge Graph (GESIS KG) [1] is a comprehensive and integrated Research Knowledge Graph (RKG) that represents metadata on social science research resources available through GESIS Search [2]. It emphasizes the semantic relationships between resources to promote reuse and enhance research discoverability. The GESIS KG includes links between various types of scientific resources — such as datasets, publications, variables, and survey instruments — as well as connections to entities like authors and social science concepts. This comprehensive approach helps overcome data fragmentation and fosters a more holistic understanding of research materials. By adhering to established standards, i.e., W3C standards, common vocabularies, and PIDs, and offering flexible access methods, i.e., different APIs, GESIS KG improves interoperability, data discoverability, captures provenance, and supports research reproducibility. It contributes to more effective Research Data Management by enabling better data integration, providing contextual information about resources, and establishing meaningful links between them. The practical applications of the GESIS KG are diverse and cater to a wide range of user needs. For example, users can explore recent publications related to specific surveys by leveraging the integrated links between resources. Researchers can analyze citation patterns, while developers can incorporate GESIS metadata into other systems using GESIS KG's standardized APIs and structured metadata — ensuring seamless integration and interoperability. In this presentation, we will emphasize accessibility and interoperability which are key objectives of the GESIS KG. While data dumps [5] provide access to the entire knowledge graph, supporting offline analysis or local integration, GESIS KG is also embedded within the GESIS Search [6], serving as a backbone to provide semantic relationships between scientific resources. Through a variety of interfaces, GESIS KG can be integrated not only within GESIS (e.g. by linking to other KGs from GESIS) but also across the NFDI and beyond. A SPARQL endpoint enables users to perform complex queries, supporting customized data extraction and analysis. This endpoint is also used to integrate GESIS KG into the 4DS Gateway [3] of NFDI4DataScience. Like other knowledge graphs within the NFDI, GESIS KG is registered in the KGI4NFDI basic service registry which makes it accessible for discovery and federation across KGs in the NFDI. Additionally, an OAI-PMH API supports metadata harvesting in standardized formats (i.e., DataCite), facilitating integration with external systems and repositories. This API will be utilized by BERD@NFDI to collect and ingest relevant metadata from GESIS into the BERD data portal [4]. Currently, the OAI-PMH API is being extended to connect GESIS KG as a metadata provider for OpenAIRE, and through it, to the European Open Science Cloud (EOSC). This integration could serve as a blueprint for connecting other NFDI data points to the EOSC EU node.