AI RESEARCH
Retrieving Floods without Floodlights: Topic Models as Binary Classifiers for Extreme Climate Events in German News
arXiv CS.CL
•
ArXi:2605.03450v1 Announce Type: new In studies of media coverage of extreme climate events, NLP methods have become indispensable for identifying relevant texts in large news databases. Still, enough annotated data to train accurate deep learning-based classifiers from scratch is often not available. Topic Models have the advantage of being both unsupervised and interpretable, but are typically used only for exploratory analysis or data characterisation.