Topic Models Conditioned on Arbitrary Features with Dirichlet-multinomial Regression

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.

Appears in Proceedings of the Twenty-Fourth Conference on Uncertainty in Artificial Intelligence (UAI2008)

References in corpus (1)

• Supervised Topic Models

Cited by in corpus (22)

• Learning Supervised Topic Models for Classification and Regression from Crowds
• Visualizing Topics with Multi-Word Expressions
• Representation Learning for Natural Language Processing
• Classification Aware Neural Topic Model and its Application on a New COVID-19 Disinformation Corpus
• Modeling Documents with Deep Boltzmann Machines
• The supervised hierarchical Dirichlet process
• Factorized Multi-Modal Topic Model
• Keyword Assisted Topic Models
• Kernel Topic Models
• Homogeneity-Based Transmissive Process to Model True and False News in Social Networks
• Item Recommendation with Evolving User Preferences and Experience
• The Nonparametric Metadata Dependent Relational Model
• Computer-Assisted Text Analysis for Social Science: Topic Models and Beyond
• The Homophily Principle in Social Network Analysis
• Friendship Prediction in Composite Social Networks
• Learning Frames from Text with an Unsupervised Latent Variable Model
• Prediction-Constrained Training for Semi-Supervised Mixture and Topic Models
• Learning Representations of Social Media Users
• Computing the Dirichlet-Multinomial Log-Likelihood Function
• Tag-Weighted Topic Model For Large-scale Semi-Structured Documents
• Variational Inference in Nonconjugate Models
• Have you tried Neural Topic Models? Comparative Analysis of Neural and Non-Neural Topic Models with Application to COVID-19 Twitter Data