We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Content Modeling Using Latent Permutations.
- Authors
Chen, Harr; Branavan, S. R. K.; Barzilay, Regina; Karger, David R.
- Abstract
We present a novel Bayesian topic model for learning discourse-level document structure. Our model leverages insights from discourse theory to constrain latent topic assignments in a way that reflects the underlying organization of document topics. We propose a global model in which both topic selection and ordering are biased to be similar across a collection of related documents. We show that this space of orderings can be effectively represented using a distribution over permutations called the Generalized Mallows Model. We apply our method to three complementary discourse-level tasks: cross-document alignment, document segmentation, and information ordering. Our experiments show that incorporating our permutation-based model in these applications yields substantial improvements in performance over previously proposed methods.
- Subjects
BAYESIAN analysis; DISCOURSE theory (Communication); PERMUTATIONS; DISCOURSE analysis; STATISTICAL decision making
- Publication
Journal of Artificial Intelligence Research, 2009, Vol 36, p129
- ISSN
1076-9757
- Publication type
Article
- DOI
10.1613/jair.2830