Model Selection for Anomaly Detection

TL;DR

This paper addresses the challenge of selecting kernels for one-class anomaly detection algorithms, proposing generalized methods and comparing their performance on synthetic and real data.

Contribution

It introduces generalized kernel selection methods for one-class classification and provides an extensive comparative analysis of these approaches.

Findings

Generalized kernel selection methods improve anomaly detection performance.

Comparison reveals strengths and weaknesses of different approaches.

Results demonstrate effectiveness on diverse datasets.

Abstract

Anomaly detection based on one-class classification algorithms is broadly used in many applied domains like image processing (e.g. detection of whether a patient is "cancerous" or "healthy" from mammography image), network intrusion detection, etc. Performance of an anomaly detection algorithm crucially depends on a kernel, used to measure similarity in a feature space. The standard approaches (e.g. cross-validation) for kernel selection, used in two-class classification problems, can not be used directly due to the specific nature of a data (absence of a second, abnormal, class data). In this paper we generalize several kernel selection methods from binary-class case to the case of one-class classification and perform extensive comparison of these approaches using both synthetic and real-world data.