Knowledge graphs are increasingly complementing AI and thus enabling open and networked science. At DIPF, the Knowledge Graph Education is being implemented as a long-term strategic project and an Open Science reference infrastructure is being established in the Education Research Portal (Fachportal Pädagogik).

Knowledge Graphs in Scientific Information

Knowledge graphs are increasingly becoming a central component of digital infrastructures. As a fact-based knowledge tool, they improve information retrieval and support AI systems through structuring. Knowledge graphs consist of a semantic network of entities. Their relationships are formally defined. The data described in this way is now machine-readable and thus forms the basis for a wide range of scientific analyses and services. Knowledge graphs thus open up new potential, especially for open and networked science. They make scientific findings comprehensible and accessible in a new way. Ideally, the data from knowledge graphs should be made available in an open and reusable form (FAIR and Open Science principle).

Knowledge Graph Education as an Open Science Evidence Infrastructure

At DIPF, the knowledge graph Education is a long-term project of the focus on Open Science ("DIPF Strategy 2027"). The aim is to integrate and link central and previously isolated databases of scientific output from educational research into the Knowledge Graph Education. A specialised reference infrastructure for educational research will be established step by step, using existing high-quality data. This infrastructure will be accessible via Education Research Portal. The Knowledge Graph Education is available for various research activities, such as data-based monitoring of educational research.

Participation and Involvement of the Communities

The research communities are actively involved in the development of the knowledge graph in various ways. Feedback from the specialist organisations is obtained, for example, through participation in specialist and user advisory boards. In addition, design workshops and needs analyses involving the research community are conducted continuously. The aim is to develop and sustainably implement the knowledge graph in a user-oriented way, closely aligned with the needs of the research communities and through continuous feedback.

Current Developments

At DIPF, the long-term strategic project is being implemented by the literature information systems unit in the Specialised Information Service Education, funded by the German Research Foundation (DFG). This department is successively creating and formalising the entity types journal, keyword, person and research data and transferring them to semantic web technologies. This involves close cooperation with the Educational Computer Science and Research Data for Education (FDB) units of the Information Centre for Education (IZB). A prototype of a calendar of events is also being developed with the German Education Server (Eduserver) and the feasibility of integrating events into the knowledge graph is being examined. In order to enable querying of the knowledge graph data, both professional (Sparql endpoint) and low-threshold access (query interface with query builder) will be provided.