ldbod: Local Density-Based Outlier Detection

Flexible procedures to compute local density-based outlier scores for ranking outliers. Both exact and approximate nearest neighbor search can be implemented, while also accommodating multiple neighborhood sizes and four different local density-based methods. It allows for referencing a random subsample of the input data or a user specified reference data set to compute outlier scores against, so both unsupervised and semi-supervised outlier detection can be implemented.

Version: 0.1.2
Depends: R (≥ 3.2.0)
Imports: stats, RANN, mnormt
Published: 2017-05-26
Author: Kristopher Williams
Maintainer: Kristopher Williams <kristopher.williams83 at gmail.com>
License: GPL-3
URL: https://github.com/kwilliams83/ldbod
NeedsCompilation: no
Materials: README
CRAN checks: ldbod results


Reference manual: ldbod.pdf
Package source: ldbod_0.1.2.tar.gz
Windows binaries: r-devel: ldbod_0.1.2.zip, r-release: ldbod_0.1.2.zip, r-oldrel: ldbod_0.1.2.zip
OS X binaries: r-release: ldbod_0.1.2.tgz, r-oldrel: ldbod_0.1.2.tgz
Old sources: ldbod archive

Reverse dependencies:

Reverse imports: OutlierDetection


Please use the canonical form https://CRAN.R-project.org/package=ldbod to link to this page.