CRAN/E | outForest

outForest

Multivariate Outlier Detection and Replacement

Installation

About

Provides a random forest based implementation of the method described in Chapter 7.1.2 (Regression model based anomaly detection) of Chandola et al. (2009) doi:10.1145/1541880.1541882. It works as follows: Each numeric variable is regressed onto all other variables by a random forest. If the scaled absolute difference between observed value and out-of-bag prediction of the corresponding random forest is suspiciously large, then a value is considered an outlier. The package offers different options to replace such outliers, e.g. by realistic values found via predictive mean matching. Once the method is trained on a reference data, it can be applied to new data.

github.com/mayer79/outForest
Bug report File report

Key Metrics

Version 1.0.1
R ≥ 3.5.0
Published 2023-05-21 503 days ago
Needs compilation? no
License GPL-2
License GPL-3
CRAN checks outForest results

Downloads

Yesterday 6 0%
Last 7 days 55 -43%
Last 30 days 264 +7%
Last 90 days 737 -6%
Last 365 days 3.514 -10%

Maintainer

Maintainer

Michael Mayer

Authors

Michael Mayer

aut / cre

Material

README
NEWS
Reference manual
Package source

Vignettes

Using 'outForest'

macOS

r-release

arm64

r-oldrel

arm64

r-release

x86_64

r-oldrel

x86_64

Windows

r-devel

x86_64

r-release

x86_64

r-oldrel

x86_64

Old Sources

outForest archive

Depends

R ≥ 3.5.0

Imports

FNN
ranger
graphics
stats
missRanger ≥ 2.1.0

Suggests

knitr
rmarkdown
testthat ≥ 3.0.0