"Dirichlet Random Forest" R package DirichletRF version 0.1.0 Implementation of the Dirichlet Random Forest algorithm for compositional response data. Supports maximum likelihood estimation ('MLE') and method-of-moments ('MOM') parameter estimation for the Dirichlet distribution. Provides two prediction strategies; averaging-based predictions (average of responses within terminal nodes) and parameter-based predictions (expected value derived from the estimated Dirichlet parameters within terminal nodes). For more details see Masoumifard, van der Westhuizen, and Gardner-Lubbe (2026, ISBN:9781032903910). Software Imports: Rcpp (>= 1.0.0), parallel LinkingTo: Rcpp Suggests: testthat (>= 3.0.0) Khaled Masoumifard <masoumifardk@yahoo.com> Comprehensive R Archive Network (CRAN) Khaled Masoumifard [aut, cre] (ORCID: <https://orcid.org/0000-0003-1921-2145>), Stephan van der Westhuizen [aut] (ORCID: <https://orcid.org/0000-0001-9469-8427>), Sugnet Lubbe [aut] (ORCID: <https://orcid.org/0000-0003-2762-9944>) GPL-3 2026-03-23 application/tgz https://CRAN.R-project.org/package=DirichletRF doi:10.32614/CRAN.package.DirichletRF