View on GitHub

stm

An R Package for the Structural Topic Model

Download this project as a .zip file Download this project as a tar.gz file

Download the Vignette

Authors: Molly Roberts, Brandon Stewart and Dustin Tingley

Please email all comments/questions to bms4 [AT] princeton.edu

News

July 31, 2018 Jan 27, 2018

Summary

The Structural Topic Model is a general framework for topic modeling with document-level covariate information. The covariates can improve inference and qualitative interpretability and are allowed to affect topical prevalence, topical content or both. The software package implements the estimation algorithms for the model and also includes tools for every stage of a standard workflow from reading in and processing raw text through making publication quality figures.

The package currently includes functionality to:

Methods Papers

Supporting Packages

Published Applications

If you have published a paper using stm that you would like to see included here please email us.