CompareLDA: A Topic Model for Document Comparison

Comparative Latent Dirichlet Allocation (CompareLDA) learns predictive topic distributions that comply with the pairwise comparison observations.

This is a GoLang project and can be assembled using standard GoLang infrastructure:

cmd/eval/chibeval.go is a Chib-style estimator of the predictive log-likelihood;
cmd/fit/rldafit.go is a CompareLDA trainer;
cmd/inf/rldainf.go is a CompareLDA predictor.

Name		Name	Last commit message	Last commit date
Latest commit History 2 Commits
cmd		cmd
ints		ints
lda		lda
model		model
umath		umath
README.md		README.md

Provide feedback