Revisiting the Exit from Nuclear Energy in Germany with NLP

Abstract

Annotation of political discourse is resource-intensive, but recent developments in NLP promise to automate complex annotation tasks. Fine-tuned transformer-based models outperform human annotators in some annotation tasks, but they require large manually annotated training datasets. In our contribution, we explore to which degree a manually annotated dataset can be automatically replicated with today's NLP methods, using unsupervised machine learning and zero- and few-shot learning.

arXiv comment23 pages, 8 figures, Accepted for publication in Zeitschrift für Diskursforschung/Journal for Discourse Studies, ISSN: 2195-867X