Knowledge (Graph) Driven Research Data Management in the Age of LLMs

One of the difficulties in general research data management lies into the fact that different scientific disciplines interpret specific concepts differently. Moreover, even if you agree about the research object to be represented, there are various different ways and principles to follow. Ontologies might help to enable FAIR research data management, i.e. to make research data findable, accessible, interoperable, and reusable. However, in the age of large language model (LLM), the effort of ontology-driven development is questioned and reevaluated. This presentation will give an introduction into research data management (RDM) in general, and in knowledge graph-basewd RDM in particular. On the example of the German National Research Data Infrastructure programme, a knowledge graph-based RDM approach is presented, accompanied by first experiments in how LLMs can improve efficiency and quality of the ontological engineering lifecycle.