Masterarbeit

The traceability and reproducibility of research results are central pillars of Open Science. In data-intensive disciplines, Jupyter notebooks – which combine text, code and visualizations in one document – frequently serve as a means to communicate computational workflows. To systematically assess the scientific quality of such computational notebooks, an existing infrastructure automatically analyzes research-related Jupyter notebooks for reproducibility and stores the results in a knowledge graph. What is missing is a lightweight but machine-actionable mechanism to communicate these results beyond the dedicated knowledge graph.

Through this thesis project, the existing automated reproducibility framework is to be expanded to include functionality for the creation, publication and processing of nanopublications. Nanopublications make it possible to share the smallest logical units of scientific statements (in triples of the form subject-predicate-object), along with their provenance, in a cryptographically signed and machine-readable format on the Semantic Web.