fastNaiveBayes: Extremely Fast Implementation of a Naive Bayes Classifier

This is an extremely fast implementation of a Naive Bayes classifier. This package is currently the only package that supports a Bernoulli distribution, a Multinomial distribution, and a Gaussian distribution, making it suitable for both binary features, frequency counts, and numerical features. Another unique feature is the support of a mix of different event models. Only numerical variables are allowed, however, categorical variables can be transformed into dummies and used with the Bernoulli distribution. This implementation offers a huge performance gain compared to the 'e1071' implementation in R. The execution times were compared on a data set of tweets and was found to be around 330 times faster. See the vignette for more details. This performance gain is only realized using a Bernoulli event model. Furthermore, the Multinomial event model implementation is even slightly faster, but incomparable since it was not implemented in 'e1071' The implementation is largely based on the paper "A comparison of event models for Naive Bayes anti-spam e-mail filtering" written by K.M. Schneider (2003) <doi:10.3115/1067807>. Any issues can be submitted to: <>.

Version: 1.0.1
Imports: Matrix, stats
Suggests: mlbench, knitr, rmarkdown, testthat
Published: 2019-03-08
Author: Martin Skogholt
Maintainer: Martin Skogholt <m.skogholt at>
License: GPL-3
NeedsCompilation: no
Materials: README NEWS
CRAN checks: fastNaiveBayes results


Reference manual: fastNaiveBayes.pdf
Vignettes: Fast Naive Bayes
Package source: fastNaiveBayes_1.0.1.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
OS X binaries: r-release: fastNaiveBayes_1.0.1.tgz, r-oldrel: fastNaiveBayes_1.0.1.tgz
Old sources: fastNaiveBayes archive


Please use the canonical form to link to this page.