@inproceedings{,
    author = {Maoro, Falk; Geierhos, Michaela},
    title = {FICODE at GermEval 2024 GerMS-Detect closed ST1 & ST2: Ensemble- and Transformer-Based Detection of Sexism and Misogyny in German Texts},
    editor = {Krenn, Brigitte; Petrak, Johann; Gross, Stephanie},
    booktitle = {Proceedings of GermEval 2024 Task 1 GerMS-Detect Workshop on Sexism Detection in German Online News Fora (GerMS-Detect 2024)},
    series = {},
    journal = {},
    address = {},
    publisher = {Association for Computational Lingustics},
    edition = {},
    year = {2024},
    isbn = {},
    volume = {},
    number =  {},
    pages = {21-25},
    url = {https://aclanthology.org/2024.germeval-2.3},
    doi = {},
    keywords = {},
    abstract = {In this paper, we present our solution for the shared task of GermEval 2024 GerMS-Detect. The joint task consists of two subtasks that we address in our solution. The texts in question may contain instances of sexism or misogyny and have been annotated in a multi-class classi- fication setting. From this setting, two tasks are derived that require different binary or multi- class classifications. We propose an ensem- ble method using multiple sequence classifica- tion models that can be applied to both sub- tasks. With respect to Subtask 1, our approach achieves an average F1 score of 0.641, and with respect to Subtask 2, our approach achieves an average Jensen-Shannon divergence of 0.354. The code is available at the following link: https://github.com/fmaoro/germeval24},
    note = {},
    institution = {Universität der Bundeswehr München, Fakultät für Informatik, INF 7 - Institut für Datensicherheit, Professur: Geierhos, Michaela},
    
    
}