# Explainable Unsupervised Anomaly Detection with Random Forest

arXiv: https://arxiv.org/abs/2504.16075 · HTML: https://mlsanity.com/paper/2504.16075/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 43 references that are in the map.
- First posted on arXiv 22 Apr 2025.

## Builds on (most cited first, 9 of 9)

- [A Unified Approach to Interpreting Model Predictions](https://mlsanity.com/paper/1705.07874/) (2017)
- [Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps](https://mlsanity.com/paper/1312.6034/) (2013)
- [ADBench: Anomaly Detection Benchmark](https://mlsanity.com/paper/2206.09426/) (2022)
- [PyOD: A Python Toolbox for Scalable Outlier Detection](https://mlsanity.com/paper/1901.01588/) (2019)
- [ECOD: Unsupervised Outlier Detection Using Empirical Cumulative Distribution Functions](https://mlsanity.com/paper/2201.00382/) (2022)
- [COPOD: Copula-Based Outlier Detection](https://mlsanity.com/paper/2009.09463/) (2020)
- [Geometry- and Accuracy-Preserving Random Forest Proximities](https://mlsanity.com/paper/2201.12682/) (2022)
- [Meta-survey on outlier and anomaly detection](https://mlsanity.com/paper/2312.07101/) (2023)
- [Geodesic Learning via Unsupervised Decision Forests](https://mlsanity.com/paper/1907.02844/) (2019)

## Built on by (most cited first, 1 of 1)

- [Interpretable Model-Aware Counterfactual Explanations for Random Forest](https://mlsanity.com/paper/2510.27397/) (2025)

## Abstract (arXiv)

We describe the use of an unsupervised Random Forest for similarity learning and improved unsupervised anomaly detection. By training a Random Forest to discriminate between real data and synthetic data sampled from a uniform distribution over the real data bounds, a distance measure is obtained that anisometrically transforms the data, expanding distances at the boundary of the data manifold. We show that using distances recovered from this transformation improves the accuracy of unsupervised anomaly detection, compared to other commonly used detectors, demonstrated over a large number of benchmark datasets. As well as improved performance, this method has advantages over other unsupervised anomaly detection methods, including minimal requirements for data preprocessing, native handling of missing data, and potential for visualizations. By relating outlier scores to partitions of the Random Forest, we develop a method for locally explainable anomaly predictions in terms of feature importance.