Explainable outlier detection through decision tree conditioning

David Cortes
1/2/2020
stat.MLcs.LG

Official code: david-cortes/outliertree (64 stars, GPL-3.0, last commit 2026-03-03).

Abstract

This work describes an outlier detection procedure (named "OutlierTree") loosely based on the GritBot software developed by RuleQuest research, which works by evaluating and following supervised decision tree splits on variables, in whose branches 1-d confidence intervals are constructed for the target variable and potential outliers flagged according to these confidence intervals. Under this logic, it's possible to produce human-readable explanations for why a given value of a variable in an observation can be considered as outlier, by considering the decision tree branch conditions along with general distribution statistics among the non-outlier observations that fell into the same branch, which can then be contrasted against the value which lies outside the CI. The supervised splits help to ensure that the generated conditions are not spurious, but rather related to the target variable and having logical breakpoints.

View on arXivView PDF

Code Implementations(1)

david-cortes/outliertreeOfficial100%
GPL-3.0

(Python, R, C++) Explainable outlier/anomaly detection through decision tree conditioning

643R, M4Jul 14, 20197 months agoGPL-3.0
anomaly-detectionoutlier-detection

Similar Papers

DTOR: Decision Tree Outlier Regressor to explain anomalies

85% match1 repo32024

Generative Adversarial Active Learning for Unsupervised Outlier Detection

83% match1 repo1322018

Extending Decision Predicate Graphs for Comprehensive Explanation of Isolation Forest

83% match2025

Interpretable Outlier Summarization

83% match2023

An empirical comparison of some outlier detection methods with longitudinal data

81% match2025

Cascade Subspace Clustering for Outlier Detection

79% match2023

Cite this paper

@article{cortes2020explainable,
  title  = {Explainable outlier detection through decision tree conditioning},
  author = {David Cortes},
  year   = {2020},
  eprint = {2001.00636},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/2001.00636}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2001.00636.svg)](https://codeofpaper.com/paper/2001.00636)

Discussion