Maria Druselmann

 

Büro A219
By appointment
Maria Druselmann is conducting doctoral research in computational linguistics, focusing on identifying the most effective language modeling techniques for generating embedding-based representations of multiword terms. These representations are intended to support the integration of complex terms into domain-specific taxonomies. This research lays the foundation for the development of advanced taxonomy expansion systems that efficiently process multiword terms. Such systems have applications in semantic search, domain-specific AI assistants, and automated knowledge base construction — all increasingly important in both academia and industry. The work involves large-scale experiments with various state-of-the-art language models and is based on the EEMWT dataset, a curated collection of semantically related multiword terms developed within the scope of this project.

📄 Related publication:
Maria Druselmann and Karin Harbusch. 2024. A Dataset of Semantically Related Multiword Terms of the Electrical Engineering Domain. In Proceedings of the 8th International Conference on Natural Language Processing and Information Retrieval (NLPIR 2024), Okayama, Japan. ACM. https://doi.org/10.1145/3711542.3711550