Ebrahim Norouzi, Dr. Anna Jacyszyn, Dr. Jörg Waitelonis, Nicole Jung

The Chemotion Electronic Lab Notebook (ELN) serves as a digitalization instrument for work in chemistry and integrates state-of-the-art Research Data Management techniques into the routine work of academic researchers, thereby improving the data that is obtained as well as the quality of information that is published.

This thesis focuses on leveraging the Chemotion Knowledge Graph (KG) to extract and enrich chemical knowledge by integrating it with other chemical databases, knowledge graphs, and large language models (LLMs). It involves mapping entities across heterogeneous data sources, applying machine learning techniques for knowledge discovery, and evaluating how LLMs can complement structured scientific data. The work contributes to the broader goal of enhancing automated scientific reasoning and supporting FAIR (Findable, Accessible, Interoperable, Reusable) data principles in chemistry.