Kernel Outlier Detection

TL;DR

Kernel Outlier Detection (KOD) is a novel, flexible, and lightweight method for high-dimensional anomaly detection that overcomes limitations of existing techniques through kernel transformation and ensemble strategies.

Contribution

The paper introduces KOD, a new outlier detection approach combining kernel transformation with ensemble search, addressing high-dimensional challenges and hyperparameter tuning issues.

Findings

Effective on small datasets with complex structures

Performs well on large benchmark datasets

Outperforms some existing methods in accuracy

Abstract

A new anomaly detection method called kernel outlier detection (KOD) is proposed. It is designed to address challenges of outlier detection in high-dimensional settings. The aim is to overcome limitations of existing methods, such as dependence on distributional assumptions or on hyperparameters that are hard to tune. KOD starts with a kernel transformation, followed by a projection pursuit approach. Its novelties include a new ensemble of directions to search over, and a new way to combine results of different direction types. This provides a flexible and lightweight approach for outlier detection. Our empirical evaluations illustrate the effectiveness of KOD on three small datasets with challenging structures, and on four large benchmark datasets.