Anomaly detection for unlabelled unit space using the Mahalanobis–Taguchi system

Masato Ohkubo, Yasushi Nagata

Research output: Contribution to journalArticle

Abstract

Alongside the progress in technology related to the Internet of Things, the Mahalanobis–Taguchi (MT) system, which is an anomaly detection technique suitable for monitoring the condition of production equipment, has attracted attention. However, with the conventional MT method, historical data acquired and accumulated from sensors and smart devices cannot be analysed appropriately. This is because very often the accumulated historical information is data not labelled as either ‘normal’ or ‘anomaly’. Therefore, in this research, we propose a procedure that enables to detect anomalies with the MT method even when learning data are categorised as ‘unlabelled’. Specifically, we introduce a process for estimating the Mahalanobis distance on the population by applying a robust method based on γ divergence in the MT method. Through numerical experiments, we show that the proposed procedure is a useful anomaly detection technique for unlabelled data. Since this implies that labelling is redundant in anomaly detection, we conclude that the practicality of the MT method can be improved.

Original languageEnglish
JournalTotal Quality Management and Business Excellence
DOIs
Publication statusPublished - 2019 Jan 1

Fingerprint

Anomaly detection
Anomaly
Labeling
Mahalanobis distance
Sensor
Numerical experiment
Internet of things
Divergence
Monitoring

Keywords

  • Gamma–divergence
  • Kullback–Leibler divergence
  • Mahalanobis–Taguchi method
  • robust estimation
  • Taguchi method

ASJC Scopus subject areas

  • Business, Management and Accounting(all)

Cite this

@article{0c2d8156e41d40d78a1e0d8e72f4746d,
title = "Anomaly detection for unlabelled unit space using the Mahalanobis–Taguchi system",
abstract = "Alongside the progress in technology related to the Internet of Things, the Mahalanobis–Taguchi (MT) system, which is an anomaly detection technique suitable for monitoring the condition of production equipment, has attracted attention. However, with the conventional MT method, historical data acquired and accumulated from sensors and smart devices cannot be analysed appropriately. This is because very often the accumulated historical information is data not labelled as either ‘normal’ or ‘anomaly’. Therefore, in this research, we propose a procedure that enables to detect anomalies with the MT method even when learning data are categorised as ‘unlabelled’. Specifically, we introduce a process for estimating the Mahalanobis distance on the population by applying a robust method based on γ divergence in the MT method. Through numerical experiments, we show that the proposed procedure is a useful anomaly detection technique for unlabelled data. Since this implies that labelling is redundant in anomaly detection, we conclude that the practicality of the MT method can be improved.",
keywords = "Gamma–divergence, Kullback–Leibler divergence, Mahalanobis–Taguchi method, robust estimation, Taguchi method",
author = "Masato Ohkubo and Yasushi Nagata",
year = "2019",
month = "1",
day = "1",
doi = "10.1080/14783363.2019.1616542",
language = "English",
journal = "Total Quality Management and Business Excellence",
issn = "1478-3363",
publisher = "Routledge",

}

TY - JOUR

T1 - Anomaly detection for unlabelled unit space using the Mahalanobis–Taguchi system

AU - Ohkubo, Masato

AU - Nagata, Yasushi

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Alongside the progress in technology related to the Internet of Things, the Mahalanobis–Taguchi (MT) system, which is an anomaly detection technique suitable for monitoring the condition of production equipment, has attracted attention. However, with the conventional MT method, historical data acquired and accumulated from sensors and smart devices cannot be analysed appropriately. This is because very often the accumulated historical information is data not labelled as either ‘normal’ or ‘anomaly’. Therefore, in this research, we propose a procedure that enables to detect anomalies with the MT method even when learning data are categorised as ‘unlabelled’. Specifically, we introduce a process for estimating the Mahalanobis distance on the population by applying a robust method based on γ divergence in the MT method. Through numerical experiments, we show that the proposed procedure is a useful anomaly detection technique for unlabelled data. Since this implies that labelling is redundant in anomaly detection, we conclude that the practicality of the MT method can be improved.

AB - Alongside the progress in technology related to the Internet of Things, the Mahalanobis–Taguchi (MT) system, which is an anomaly detection technique suitable for monitoring the condition of production equipment, has attracted attention. However, with the conventional MT method, historical data acquired and accumulated from sensors and smart devices cannot be analysed appropriately. This is because very often the accumulated historical information is data not labelled as either ‘normal’ or ‘anomaly’. Therefore, in this research, we propose a procedure that enables to detect anomalies with the MT method even when learning data are categorised as ‘unlabelled’. Specifically, we introduce a process for estimating the Mahalanobis distance on the population by applying a robust method based on γ divergence in the MT method. Through numerical experiments, we show that the proposed procedure is a useful anomaly detection technique for unlabelled data. Since this implies that labelling is redundant in anomaly detection, we conclude that the practicality of the MT method can be improved.

KW - Gamma–divergence

KW - Kullback–Leibler divergence

KW - Mahalanobis–Taguchi method

KW - robust estimation

KW - Taguchi method

UR - http://www.scopus.com/inward/record.url?scp=85065793162&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85065793162&partnerID=8YFLogxK

U2 - 10.1080/14783363.2019.1616542

DO - 10.1080/14783363.2019.1616542

M3 - Article

AN - SCOPUS:85065793162

JO - Total Quality Management and Business Excellence

JF - Total Quality Management and Business Excellence

SN - 1478-3363

ER -