Unsupervised Learning Based Definition of Microbial Rare Biosphere R package ulrb version 0.1.8 A tool to define the rare biosphere. 'ulrb' solves the problem of the definition of rarity by replacing arbitrary thresholds with an unsupervised machine learning algorithm (partitioning around medoids, or k-medoids). This algorithm works for any type of microbiome data, provided there is an abundance table. This method also works for non-microbiome data. Software Depends: R (>= 2.10) Imports: cluster, dplyr, ggplot2, purrr, rlang, stats, tidyr, clusterSim, gridExtra Suggests: knitr, rmarkdown, stringr, testthat (>= 3.0.0), vegan Francisco Pascoal <fpascoal1996@gmail.com> Comprehensive R Archive Network (CRAN) Francisco Pascoal [aut, cre] (ORCID: <https://orcid.org/0000-0003-2315-358X>), Paula Branco [aut] (ORCID: <https://orcid.org/0000-0002-9917-3694>), Luís Torgo [aut] (ORCID: <https://orcid.org/0000-0002-6892-8871>), Rodrigo Costa [aut] (ORCID: <https://orcid.org/0000-0002-5932-4101>), Catarina Magalhães [aut] (ORCID: <https://orcid.org/0000-0001-9576-2398>) GPL (>= 3) 2025-07-07 application/tgz https://CRAN.R-project.org/package=ulrb doi:10.32614/CRAN.package.ulrb