Sparse random projection isolation forest for outlier detection

Xu Tan,Jiawei Yang,Susanto Rahardja
DOI: https://doi.org/10.1016/j.patrec.2022.09.015
IF: 4.757
2022-11-01
Pattern Recognition Letters
Abstract: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-projection-based 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.
computer science, artificial intelligence
What problem does this paper attempt to address?