Johanna Rockstroh

• Knowledge representation and ontologies

• Formal semantics

• Machine learning models in Natural Language Processing

• Human Computer Interaction (HCI)

Possible thesis projects:

• enabling chatbots to meaningfully use different kinds of media (e.g photos, videos, GIFs):
  • topic modelling/recognition
  • sentiment analysis
  • semantic analysis
  • text to knowledge graph parsing
• neuro-symbolic methods to harvest language samples from (parallel) corpora:
  • frame parsing
  • searching knowledge graphs
  • enhancing knowledge graphs
  • speech act classification

Johanna holds a Master's degree in Foreign Language Linguistics and studied Computational Linguistics at the University of Potsdam. Her research interests are in natural language processing (NLP) and human-robot interaction.

She is currently working on methods for extracting semantic frames in the context of parallel corpora.

Publications:

ORCID

Selected publications:

Rockstroh, J., D’Ippolito, G., Lazzari, N., Oudshoorn, A. M., Purohit, D., Raoufi, E., & Rudolph, S. (2023). A is the B of C: (Semi)-Automatic Creation of Vossian Antonomasias. In L.-A. Kaffee, S. Razniewski, K. Alghamdi, & H. Arnaout (Chairs), Proceedings of the Wikidata Workshop 2023, co-located with 22nd International Semantic Web Conference (ISWC 2023).

Avanesi, V., Rockstroh, J., Mildner, T., Zargham, N., Reicherts, L., Friehs, M. A., Kontogiorgos, D., Wenig, N., & Malaka, R. (2023). From C-3PO to HAL: Opening The Discourse About The Dark Side of Multi-Modal Social Agents. In M. Lee, C. Munteanu, M. Porcheron, J. Trippas, & S. T. Völkel (Eds.), Proceedings of the 5th International Conference on Conversational User Interfaces (pp. 1–7).

• 2021 Hans-Jürgen Bachorski-award for Code-Switching auf Twitter: Ein Vergleich zwischen El Paso und Gibraltar