KentonWhite/nmftm: Creates a Topic Model of short texts using Non-Negative Matrix Factorization

Topic Modeling is usually done with Latent Dirichlet Allocation. This works great for long documents but seems to fail for short documents, like Tweets. In this package, Topic Modeling is implemented using Non-Negative Matrix Factorization (NMF). NMF is an altnernate that produced stronger topci cohesion when each document is short.

install.packages("remotes")
remotes::install_github("KentonWhite/nmftm")