Paper: Learning Topics and Positions from Debatepedia

ACL ID D13-1191
Title Learning Topics and Positions from Debatepedia
Venue Conference on Empirical Methods in Natural Language Processing
Session Main Conference
Year 2013
Authors

We explore Debatepedia, a community- authored encyclopedia of sociopolitical de- bates, as evidence for inferring a low- dimensional, human-interpretable representa- tion in the domain of issues and positions. We introduce a generative model positing latent topics and cross-cutting positions that gives special treatment to person mentions and opin- ion words. We evaluate the resulting repre- sentation?s usefulness in attaching opinionated documents to arguments and its consistency with human judgments about positions.