emnlp15-dim4auc

This repository contains the code used to perform the classification experiments described in section 4.2 of the paper:

Judith Eckle-Kohler, Roland Kluge and Iryna Gurevych. On the Role of Discourse Markers for Discriminating Claims and Premises in Argumentative Discourse. In: Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing (EMNLP), p. 2249-2255, September 2015.

For more information about downloads see also: https://www.ukp.tu-darmstadt.de/data/argumentation-mining/argument-annotated-news-articles/

Package structure

• the package experiment contains the runnable classification experiments which make use of the DKPro TC framework.
• the package explore contains code used for the extraction of data-driven features from the German Tiger corpus. It makes use of DKPro Core.

Howto

• to obtain the annotated dataset, please contact [email protected] or the primary contact of the argumentation mining group at UKP, see https://www.ukp.tu-darmstadt.de/research/research-areas/argumentation-mining
• runnable classification experiments are in the package experiment; in order to run them, you have to modify the paths to corpus data and results data according to your system and environment