Learning Topics and Positions from Debatepedia

Author

Swapna GOTTOPATI, Singapore Management UniversityFollow
Minghui QIU, Singapore Management UniversityFollow
Yanchuan SIM, Carnegie Mellon University
Jing JIANG, Singapore Management UniversityFollow
Noah SMITH, Carnegie Mellon University

Publication Type

Conference Proceeding Article

Version

acceptedVersion

Publication Date

10-2013

Abstract

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.

Discipline

Databases and Information Systems | Numerical Analysis and Scientific Computing

Publication

Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing, Seattle, Washington, 18-21 October 2013

First Page

1858

Last Page

1868

Publisher

ACL

City or Country

Stroudsburg, PA

Citation

GOTTOPATI, Swapna; QIU, Minghui; SIM, Yanchuan; JIANG, Jing; and SMITH, Noah. Learning Topics and Positions from Debatepedia. (2013). Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing, Seattle, Washington, 18-21 October 2013. 1858-1868.
Available at: https://ink.library.smu.edu.sg/sis_research/2059

Copyright Owner and License

LARC

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 4.0 International License.

Additional URL

http://www.aclweb.org/anthology/D13-1191