From 016cabf7bee007fb580d355113de8aa13d0f0a2b Mon Sep 17 00:00:00 2001 From: pimi Date: Mon, 10 Sep 2018 13:21:38 +0200 Subject: [PATCH] gnu: Add r-abodoutlier. * gnu/packages/cran.scm (r-abodoutlier): New variable. Signed-off-by: Ricardo Wurmus --- gnu/packages/cran.scm | 29 +++++++++++++++++++++++++++++ 1 file changed, 29 insertions(+) diff --git a/gnu/packages/cran.scm b/gnu/packages/cran.scm index 439cfcb014..0c6a4908e8 100644 --- a/gnu/packages/cran.scm +++ b/gnu/packages/cran.scm @@ -5822,3 +5822,32 @@ Using high-dimensional datasets to measure a subject's overall level of abnormality as compared to a reference population is often needed in outcomes research.") (license license:expat))) + +(define-public r-abodoutlier + (package + (name "r-abodoutlier") + (version "0.1") + (source + (origin + (method url-fetch) + (uri (cran-uri "abodOutlier" version)) + (sha256 + (base32 "1pvhgxmh23br84r0fbmv7g53z2427birdja96a67vqgz18r3fdvj")))) + (properties `((upstream-name . "abodOutlier"))) + (build-system r-build-system) + (propagated-inputs + `(("r-cluster" ,r-cluster))) + (home-page "https://cran.r-project.org/web/packages/abodOutlier/") + (synopsis "Angle-based outlier detection") + (description + "This package performs angle-based outlier detection on a given data +frame. It offers three methods to process data: +@enumerate +@item full but slow implementation using all the data that has cubic + complexity; +@item a fully randomized method; +@item a method using k-nearest neighbours. +@end enumerate +These algorithms are well suited for high dimensional data outlier +detection.") + (license license:expat)))