Anomaly Detection with the Voronoi Diagram Evolutionary Algorithm

TL;DR

This paper introduces VorEAl, a novel evolutionary algorithm that uses Voronoi diagrams to improve anomaly detection by effectively partitioning data space and handling unseen data regions.

Contribution

VorEAl is the first evolutionary algorithm leveraging Voronoi diagrams for anomaly detection, optimizing classification and data space representation simultaneously.

Findings

VorEAl outperforms similar methods in anomaly detection tasks.

It effectively identifies abnormal regions including unseen data areas.

The approach demonstrates robustness across different datasets.

Abstract

This paper presents the Voronoi diagram-based evolutionary algorithm (VorEAl). VorEAl partitions input space in abnormal/normal subsets using Voronoi diagrams. Diagrams are evolved using a multi-objective bio-inspired approach in order to conjointly optimize classification metrics while also being able to represent areas of the data space that are not present in the training dataset. As part of the paper VorEAl is experimentally validated and contrasted with similar approaches.