Time domain graph-based anomaly detection approach applied to a real industrial problem

Carregando...
Imagem de Miniatura

Data

Título da Revista

ISSN da Revista

Título de Volume

Editor

Universidade Federal de Minas Gerais

Descrição

Tipo

Artigo de periódico

Título alternativo

Primeiro orientador

Membros da banca

Resumo

Detecting anomalies in industrial processes is a critical task. Prior fault detection can reduce company costs, and most importantly, may prevent accidents and environmental damage. Anomaly detection can be treated as drift detection, as both aims at identifying changes in data that happen unexpectedly over time. In this paper, a graph-based approach to anomaly detection is presented. Based on graph theory and set coverage, the method brings results similar to deep autoencoders, without the need to extensively set parameters.

Abstract

Assunto

Aprendizado de máquina

Palavras-chave

Anomaly detection, Gabriel graph, Dominating set, Deep autoenconders, Machine learning, A graph-based approach to anomaly detection is presented. This work shows that dominating set incorporates statistical properties of a data set. The concept of a space coverage to encapsulate a data set, using dominating sets, is presented. The new approach brings results similar to deep autoencoders, without the need to extensively set parameters.

Citação

Curso

Endereço externo

https://www.sciencedirect.com/science/article/pii/S0166361522001117

Avaliação

Revisão

Suplementado Por

Referenciado Por