Revisiting the Exit from Nuclear Energy in Germany with NLP
This work addresses the resource-intensive annotation of political discourse for researchers, but it is incremental as it applies existing NLP methods to a specific domain.
The study tackled the problem of replicating a manually annotated political discourse dataset on Germany's nuclear energy exit using NLP methods, achieving results through unsupervised and few-shot learning techniques.
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.