German Argument Similarity dataset (GerArgSimilarity)

This is the German-language argument similarity dataset accompanying our paper titled Argument Similarity Assessment in German for Intelligent Tutoring: Crowdsourced Dataset and First Experiments (to be presented at LREC 2022). It consists of 2940 argumentative pairs of text snippets in German which are annotated for argument similarity on a scale from 0 (no similarity) to 4 (high similarity). The snippets cover arguments on 3 discussion topics. Please refer to the paper for details.