Package: songR
Type: Package
Title: Self-Organizing Nebulous Growths for Dimensionality Reduction
Version: 0.1.0
Authors@R: c(
    person("R.", "Heller", role = c("aut", "cre", "cph"),
           email = "raban.heller@uni-ulm.de",
           comment = c(ORCID = "0000-0001-8006-9742")),
    person("Damith A.", "Senanayake", role = "cph",
           comment = "Author of the original SONG algorithm and its BSD-3-Clause Python implementation; see inst/COPYRIGHTS"))
Description: A SONG-inspired tool for nonlinear dimensionality
        reduction and data visualization in R, implementing the
        Self-Organizing Nebulous Growths (SONG) algorithm of Senanayake
        et al. (2021) <doi:10.1109/TNNLS.2020.3023941> natively in R
        and C++ and staying as close to the reference implementation as
        is feasible across the two ecosystems. SONG is a parametric
        method that supports incremental addition of new data to
        existing visualizations without reinitialization, handles both
        homogeneous and heterogeneous data increments, and is robust to
        noise and highly mixed clusters. Clustering quality matches the
        reference closely (validated by reference-parity tests); the
        default visualization is close in global structure but not
        identical in absolute layout. Includes an interactive Shiny
        application for comparing SONG, t-SNE, and UMAP visualizations
        side-by-side.
License: MIT + file LICENSE
Encoding: UTF-8
Language: en-US
LazyData: true
Roxygen: list(markdown = TRUE)
Depends: R (>= 4.0.0)
Imports: cli, digest, Rcpp (>= 1.0.0), rlang
Suggests: Matrix, ggplot2, viridis, shiny, Rtsne, uwot, aricode,
        testthat (>= 3.0.0), withr, knitr, rmarkdown
LinkingTo: Rcpp, RcppArmadillo
Config/testthat/edition: 3
Config/Needs/website: pkgdown, CTTIR/themakR
VignetteBuilder: knitr
URL: https://cttir.github.io/songR/, https://github.com/cttir/songR
BugReports: https://github.com/cttir/songR/issues
Config/roxygen2/version: 8.1.0
Repository: https://cttir.r-universe.dev
Date/Publication: 2026-10-28 07:29:20 UTC
RemoteUrl: https://github.com/CTTIR/songR
RemoteRef: HEAD
RemoteSha: b8572dd53f2b52d1ac32c2dd31995d38552e432e
NeedsCompilation: yes
Packaged: 2026-10-29 10:32:38 UTC; root
Author: R. Heller [aut, cre, cph] (ORCID:
    <https://orcid.org/0000-0001-8006-9742>),
  Damith A. Senanayake [cph] (Author of the original SONG algorithm and
    its BSD-3-Clause Python implementation; see inst/COPYRIGHTS)
Maintainer: R. Heller <raban.heller@uni-ulm.de>