Sparse random projection isolation forest for outlier detection

Xu Tan

doi:10.1016/j.patrec.2022.09.015

Sparse random projection isolation forest for outlier detection

2022-11-01

25

PRE

AI

DOI：10.1016/j.patrec.2022.09.015

Original

Original request for help

Save

Abstract

En 中文

Isolation Forest has a low computational complexity, hence has been widely applied to detect outliers in large-scale data. However, it suffers from the artifacts caused by the hyperplanes chosen, thereby failing to detect outliers in some specific regions. To tackle this problem, we propose the random-projectionbased Isolation Forest, which works in two steps. First, we transform the data using the random projection technique. Then, we employ the Isolation Forest to identify outliers using the transformed data. Experimental results show that the proposed methods outperform 12 state-of-the-art outlier detectors.(c) 2022 Published by Elsevier B.V.

Keywords:

Outlier detection

Anomaly detection

Isolation forest

Random projection

Sparse random projection

Sparse random projection isolation forest for outlier detection

Abstract

Journal

Organization