sbert: Topic Modeling with 'Sentence-BERT' Embeddings

Builds semantic topic models from 'Sentence-BERT' embeddings (Reimers and Gurevych, 2019) <doi:10.18653/v1/D19-1410>. Sentence embeddings are computed in several modes with models run on the Open Neural Network Exchange ('ONNX') Runtime, including multilingual and long-context encoders and a fast static model. Topics are labeled with class-based term weighting that can be retuned without refitting. Candidate topic counts and document granularities (whole documents, sentences, clauses, or phrases) can be compared side by side, and fitted models assign topics to new documents, give soft memberships and per-document topic mixtures. The package offers coherence and diversity metrics and visualizations.

Version:	0.5.8
Depends:	R (≥ 4.1.0)
Imports:	digest, grDevices, graphics, onnxr (≥ 0.1.2), parallel, tok (≥ 0.2.3), stats, tools, utils
Suggests:	SnowballC, jsonlite, knitr, rmarkdown, testthat (≥ 3.0.0), withr
Published:	2026-10-21
DOI:	10.32614/CRAN.package.sbert
Author:	Mohammed Saqr [aut, cph], Sonsoles López-Pernas [aut, cre, cph]
Maintainer:	Sonsoles López-Pernas <sonsoles.lopez at uef.fi>
BugReports:	https://github.com/sonsoleslp/sbert/issues
License:	MIT + file LICENSE
Copyright:	see file COPYRIGHTS
URL:	https://sonsoles.me/sbert/, https://github.com/sonsoleslp/sbert
NeedsCompilation:	no
SystemRequirements:	ONNX Runtime (>= 1.25.1)
Language:	en-US
Materials:	README, NEWS
CRAN checks:	sbert results

Documentation:

Reference manual:	sbert.html , sbert.pdf
Vignettes:	Semantic Topics in AI-Powered Mathematics Feedback (source, R code)

Downloads:

Package source:	sbert_0.5.8.tar.gz
Windows binaries:	r-devel: not available, r-release: not available, r-oldrel: sbert_0.5.8.zip
macOS binaries:	r-release (arm64): sbert_0.5.8.tgz, r-oldrel (arm64): sbert_0.5.8.tgz, r-release (x86_64): sbert_0.5.8.tgz, r-oldrel (x86_64): sbert_0.5.8.tgz

Linking:

Please use the canonical form https://CRAN.R-project.org/package=sbert to link to this page.