fastTopics: Fast Algorithms for Fitting Topic Models and Non-Negative Matrix Factorizations to Count Data

Implements fast, scalable optimization algorithms for fitting topic models ("grade of membership" models) and non-negative matrix factorizations to count data. The methods exploit the special relationship between the multinomial topic model (also, "probabilistic latent semantic indexing") and Poisson non-negative matrix factorization. The package provides tools to compare, annotate and visualize model fits, including functions to efficiently create "structure plots" and identify key features in topics. The 'fastTopics' package is a successor to the 'CountClust' package.

Version: 0.6-142
Depends: R (≥ 3.3.0)
Imports: graphics, utils, methods, stats, Matrix, MCMCpack, quadprog, irlba, dplyr, Rtsne, uwot, ashr, Rcpp (≥ 1.0.1), RcppParallel (≥ 4.4.1), parallel, progress, pbapply, ggplot2 (≥ 3.3.0), ggrepel (≥ 0.9.0), cowplot, plotly, htmlwidgets
LinkingTo: Rcpp, RcppParallel, RcppArmadillo
Suggests: NNLM, Ternary, testthat, knitr, rmarkdown
Published: 2022-09-06
Author: Peter Carbonetto [aut, cre], Kevin Luo [aut], Kushal Dey [aut], Joyce Hsiao [ctb], Abhishek Sarkar [ctb], Anthony Hung [ctb], Xihui Lin [ctb], Paul C. Boutros [ctb], Minzhe Wang [ctb], Tracy Ke [ctb], Matthew Stephens [aut]
Maintainer: Peter Carbonetto <peter.carbonetto at>
License: BSD_2_clause + file LICENSE
Copyright: inst/COPYRIGHTS
fastTopics copyright details
NeedsCompilation: yes
SystemRequirements: GNU make
Citation: fastTopics citation info
Materials: README
CRAN checks: fastTopics results


Reference manual: fastTopics.pdf
Vignettes: Relationship between NMF and topic modeling
Topic modeling vs. clustering


Package source: fastTopics_0.6-142.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
macOS binaries: r-release (arm64): fastTopics_0.6-142.tgz, r-oldrel (arm64): fastTopics_0.6-142.tgz, r-release (x86_64): fastTopics_0.6-142.tgz, r-oldrel (x86_64): fastTopics_0.6-142.tgz
Old sources: fastTopics archive


Please use the canonical form to link to this page.