Open Access
Translator Disclaimer
March 2010 Hierarchical relational models for document networks
Jonathan Chang, David M. Blei
Ann. Appl. Stat. 4(1): 124-150 (March 2010). DOI: 10.1214/09-AOAS309

Abstract

We develop the relational topic model (RTM), a hierarchical model of both network structure and node attributes. We focus on document networks, where the attributes of each document are its words, that is, discrete observations taken from a fixed vocabulary. For each pair of documents, the RTM models their link as a binary random variable that is conditioned on their contents. The model can be used to summarize a network of documents, predict links between them, and predict words within them. We derive efficient inference and estimation algorithms based on variational methods that take advantage of sparsity and scale with the number of links. We evaluate the predictive performance of the RTM for large networks of scientific abstracts, web documents, and geographically tagged news.

Citation

Download Citation

Jonathan Chang. David M. Blei. "Hierarchical relational models for document networks." Ann. Appl. Stat. 4 (1) 124 - 150, March 2010. https://doi.org/10.1214/09-AOAS309

Information

Published: March 2010
First available in Project Euclid: 11 May 2010

zbMATH: 1189.62191
MathSciNet: MR2758167
Digital Object Identifier: 10.1214/09-AOAS309

Keywords: Mixed-membership models , network models , text analysis , variational methods

Rights: Copyright © 2010 Institute of Mathematical Statistics

JOURNAL ARTICLE
27 PAGES


SHARE
Vol.4 • No. 1 • March 2010
Back to Top