Anchored Correlation Explanation: Topic Modeling with Minimal Domain Knowledge
Published
2017-12-02
Ryan J. Gallagher
,
Kyle Reing
,
David Kale
,
Greg Ver Steeg
Ryan J. Gallagher
University of Vermont
Information Sciences Institute, University of Southern California
Kyle Reing
Information Sciences Institute, University of Southern California
David Kale
Information Sciences Institute, University of Southern California
Greg Ver Steeg
Information Sciences Institute, University of Southern California
Abstract
While generative models such as Latent Dirichlet Allocation (LDA) have proven fruitful in topic modeling, they often require detailed assumptions and careful specification of hyperparameters. Such model complexity issues only compound when trying to generalize generative models to incorporate human input. We introduce Correlation Explanation (CorEx), an alternative approach to topic modeling that does not assume an underlying generative model, and instead learns maximally informative topics through an information-theoretic framework. This framework naturally generalizes to hierarchical and semi-supervised extensions with no additional modeling assumptions. In particular, word-level domain knowledge can be flexibly incorporated within CorEx through anchor words, allowing topic separability and representation to be promoted with minimal human intervention. Across a variety of datasets, metrics, and experiments, we demonstrate that CorEx produces topics that are comparable in quality to those produced by unsupervised and semi-supervised variants of LDA.
PDF (presented at NAACL 2018)
,