Date of Original Version
Proceedings of Conference e on Empirical Methods in Natural Language Processing (EMNLP)
Copyright 2013 Association for Computational Linguistics
Abstract or Description
We explore Debatepedia, a communityauthored encyclopedia of sociopolitical debates, as evidence for inferring a lowdimensional, human-interpretable representation 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 opinion words. We evaluate the resulting representation’s usefulness in attaching opinionated documents to arguments and its consistency with human judgments about positions.
Proceedings of Conference e on Empirical Methods in Natural Language Processing (EMNLP), 1858-1868.