Hands-on-Workshop: Research Data Management with Machine Training Support

The presentation explains how research data within the Cluster of Excellence Understanding Written Artefacts (UWA) is managed and published via the University of Hamburg’s Research Data Repository (RDR). Researchers are encouraged not just to upload files but to present their data in ways that make it transparent, reusable, and compliant with FAIR (Findable, Accessible, Interoperable, Reusable) principles. A key tool in this work are CSMC files, archive packages similar to DOCX, which include project-specific views of the data. These CSMC files can be submitted to RDR, where they receive persistent identifiers (DOIs) and become accessible online. Examples from UWA projects illustrate how datasets appear in the repository with customized views rather than simple directory listings. A CSMC App also allows scholars to view and interact with their data locally, even without internet access. When a dataset is submitted, RDR automatically generates a “View Data” button so anyone can see the data presentation defined by the creators. Individual data records within these datasets can be cited directly using fragment identifiers in DOIs. The page highlights the importance of data persistence and versioning in making scholarly research reproducible and citable. Furthermore, we discuss research data from RDR and visualized as described above can be made available for fine-tuning generalized pretrained transformers to easier carry out humanities research tasks.