outliertree: Explainable Outlier Detection Through Decision Tree Conditioning

Will try to fit decision trees that try to "predict" values for each column based on the values of each other column. Along the way, each time a split is evaluated, it will take the observations that fall into each branch as a homogeneous cluster in which it will search for outliers in the 1-d distribution of the column being predicted. Outliers are determined according to confidence intervals in this 1-d distribution, and need to have a large gap with respect to the next observation in sorted order to be flagged as outliers. Since outliers are searched for in a decision tree branch, it will know the conditions that make it a rare observation compared to others that meet the same conditions, and the conditions will always be correlated with the target variable (as it's being predicted from them). Full procedure is described in Cortes (2020) <arXiv:2001.00636>. Loosely based on the 'GritBot' <https://www.rulequest.com/gritbot-info.html> software.

Version: 1.3.0
Depends: R (≥ 3.5.0)
Imports: Rcpp (≥ 1.0.1)
LinkingTo: Rcpp, Rcereal
Published: 2020-06-10
Author: David Cortes
Maintainer: David Cortes <david.cortes.rivera at gmail.com>
BugReports: https://github.com/david-cortes/outliertree/issues
License: GPL (≥ 3)
URL: https://github.com/david-cortes/outliertree
NeedsCompilation: yes
CRAN checks: outliertree results


Reference manual: outliertree.pdf
Package source: outliertree_1.3.0.tar.gz
Windows binaries: r-devel: outliertree_1.3.0.zip, r-release: outliertree_1.3.0.zip, r-oldrel: outliertree_1.3.0.zip
macOS binaries: r-release: outliertree_1.3.0.tgz, r-oldrel: outliertree_1.3.0.tgz
Old sources: outliertree archive

Reverse dependencies:

Reverse suggests: isotree


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