Towards segment-based recognition of argumentation structure in short texts

Despite recent advances in discourse pars-ing and causality detection, the automatic recognition of argumentation structure of authentic texts is still a very challeng-ing task. To approach this problem, we collected a small corpus of German mi-crotexts in a text generation experiment, resulting in texts that are authentic but of controlled linguistic and rhetoric com-plexity. We show that trained annotators can determine the argumentation struc-ture on these microtexts reliably. We ex-periment with different machine learning approaches for automatic argumentation structure recognition on various levels of granularity of the scheme. Given the com-plex nature of such a discourse under-standing tasks, the first results presented here are promising, but invite for further investigation.