Knowledge Discovery by Accuracy Maximization R package KODAMA version 3.3 A self-guided, weakly supervised learning algorithm for feature extraction from noisy and high-dimensional data. It facilitates the identification of patterns that reflect underlying group structures across all samples in a dataset. The method incorporates a novel strategy to integrate spatial information, improving the clarity of results in spatially resolved data. Software Depends: R (>= 2.10.0), stats, Rtsne, umap Imports: Rcpp (>= 0.12.4), Rnanoflann, methods, Matrix LinkingTo: Rcpp, RcppArmadillo, Rnanoflann, Matrix Suggests: rgl, knitr, rmarkdown, testthat (>= 3.0.0) Stefano Cacciatore <tkcaccia@gmail.com> Comprehensive R Archive Network (CRAN) Stefano Cacciatore [aut, trl, cre] (ORCID: <https://orcid.org/0000-0001-7052-7156>), Leonardo Tenori [aut] (ORCID: <https://orcid.org/0000-0001-6438-059X>) GPL (>= 2) 2026-03-17 application/tgz https://CRAN.R-project.org/package=KODAMA doi:10.32614/CRAN.package.KODAMA