Topic Models Conditioned on Arbitrary Features with Dirichlet-multinomial Regression
Information Retrieval
2012-06-18 v1 Methodology
Abstract
Although fully generative models have been successfully used to model the contents of text documents, they are often awkward to apply to combinations of text data and document metadata. In this paper we propose a Dirichlet-multinomial regression (DMR) topic model that includes a log-linear prior on document-topic distributions that is a function of observed features of the document, such as author, publication venue, references, and dates. We show that by selecting appropriate features, DMR topic models can meet or exceed the performance of several previously published topic models designed for specific data.
Keywords
Cite
@article{arxiv.1206.3278,
title = {Topic Models Conditioned on Arbitrary Features with Dirichlet-multinomial Regression},
author = {David Mimno and Andrew McCallum},
journal= {arXiv preprint arXiv:1206.3278},
year = {2012}
}
Comments
Appears in Proceedings of the Twenty-Fourth Conference on Uncertainty in Artificial Intelligence (UAI2008)