Home > Research > Publications & Outputs > A linear time method for the detection of colle...

Links

Text available via DOI:

View graph of relations

A linear time method for the detection of collective and point anomalies

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

A linear time method for the detection of collective and point anomalies. / Fisch, Alexander T. M.; Eckley, Idris A.; Fearnhead, Paul.

In: Statistical Analysis and Data Mining, Vol. 15, No. 4, 31.08.2022, p. 494-508.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

APA

Vancouver

Fisch ATM, Eckley IA, Fearnhead P. A linear time method for the detection of collective and point anomalies. Statistical Analysis and Data Mining. 2022 Aug 31;15(4):494-508. Epub 2022 Jun 3. doi: 10.1002/sam.11586

Author

Bibtex

@article{7a3a43119fb44e61bf5df0efd8bbf44b,
title = "A linear time method for the detection of collective and point anomalies",
abstract = "Abstract: The challenge of efficiently identifying anomalies in data sequences is an important statistical problem that now arises in many applications. Although there has been substantial work aimed at making statistical analyses robust to outliers, or point anomalies, there has been much less work on detecting anomalous segments, or collective anomalies, particularly in those settings where point anomalies might also occur. In this article, we introduce collective and point anomalies (CAPA), a computationally efficient approach that is suitable when collective anomalies are characterized by either a change in mean, variance, or both, and distinguishes them from point anomalies. Empirical results show that CAPA has close to linear computational cost as well as being more accurate at detecting and locating collective anomalies than other approaches. We demonstrate the utility of CAPA through its ability to detect exoplanets from light curve data from the Kepler telescope and its capacity to detect machine faults from temperature data.",
keywords = "RESEARCH ARTICLE, RESEARCH ARTICLES, dynamic programming, epidemic changepoints, exoplanets, Numenta Anomaly Benchmark, outliers, robust statistics",
author = "Fisch, {Alexander T. M.} and Eckley, {Idris A.} and Paul Fearnhead",
year = "2022",
month = aug,
day = "31",
doi = "10.1002/sam.11586",
language = "English",
volume = "15",
pages = "494--508",
journal = "Statistical Analysis and Data Mining",
issn = "1932-1864",
publisher = "John Wiley and Sons Inc.",
number = "4",

}

RIS

TY - JOUR

T1 - A linear time method for the detection of collective and point anomalies

AU - Fisch, Alexander T. M.

AU - Eckley, Idris A.

AU - Fearnhead, Paul

PY - 2022/8/31

Y1 - 2022/8/31

N2 - Abstract: The challenge of efficiently identifying anomalies in data sequences is an important statistical problem that now arises in many applications. Although there has been substantial work aimed at making statistical analyses robust to outliers, or point anomalies, there has been much less work on detecting anomalous segments, or collective anomalies, particularly in those settings where point anomalies might also occur. In this article, we introduce collective and point anomalies (CAPA), a computationally efficient approach that is suitable when collective anomalies are characterized by either a change in mean, variance, or both, and distinguishes them from point anomalies. Empirical results show that CAPA has close to linear computational cost as well as being more accurate at detecting and locating collective anomalies than other approaches. We demonstrate the utility of CAPA through its ability to detect exoplanets from light curve data from the Kepler telescope and its capacity to detect machine faults from temperature data.

AB - Abstract: The challenge of efficiently identifying anomalies in data sequences is an important statistical problem that now arises in many applications. Although there has been substantial work aimed at making statistical analyses robust to outliers, or point anomalies, there has been much less work on detecting anomalous segments, or collective anomalies, particularly in those settings where point anomalies might also occur. In this article, we introduce collective and point anomalies (CAPA), a computationally efficient approach that is suitable when collective anomalies are characterized by either a change in mean, variance, or both, and distinguishes them from point anomalies. Empirical results show that CAPA has close to linear computational cost as well as being more accurate at detecting and locating collective anomalies than other approaches. We demonstrate the utility of CAPA through its ability to detect exoplanets from light curve data from the Kepler telescope and its capacity to detect machine faults from temperature data.

KW - RESEARCH ARTICLE

KW - RESEARCH ARTICLES

KW - dynamic programming

KW - epidemic changepoints

KW - exoplanets

KW - Numenta Anomaly Benchmark

KW - outliers

KW - robust statistics

U2 - 10.1002/sam.11586

DO - 10.1002/sam.11586

M3 - Journal article

VL - 15

SP - 494

EP - 508

JO - Statistical Analysis and Data Mining

JF - Statistical Analysis and Data Mining

SN - 1932-1864

IS - 4

ER -