Autonomous anomaly detection

Xiaowei Gu; Plamen Parvanov Angelov

doi:10.1109/EAIS.2017.7954831

Autonomous anomaly detection

Xiaowei Gu
, Plamen Parvanov Angelov

Lancaster University

Research output: Chapter in Book/Report/Conference proceeding › Conference Proceeding (ISBN)

22 Citations (Scopus)

Abstract

In this paper, a new approach for autonomous anomaly detection is introduced within the Empirical Data Analytics (EDA) framework. This approach is fully data-driven and free from thresholds. Employing the nonparametric EDA estimators, the proposed approach can autonomously detect anomalies in an objective way based on the mutual distribution and ensemble properties of the data. The proposed approach firstly identifies the potential anomalies based on two EDA criteria, and then, partitions them into shape-free, non-parametric data clouds. Finally, it identifies the anomalies in regards to each data cloud (locally). Numerical examples based on synthetic and benchmark datasets demonstrate the validity and efficiency of the proposed approach.

Original language	English
Title of host publication	IEEE Conference on Evolving and Adaptive Intelligent Systems 2017
Subtitle of host publication	Conference Proceedings
Editors	Igor Skrjanc, Saso Blazic
Publisher	Institute of Electrical and Electronics Engineers
ISBN (Print)	9781509064441
DOIs	https://doi.org/10.1109/EAIS.2017.7954831
Publication status	E-pub ahead of print - 22 Jun 2017
Externally published	Yes

Publication series

Name	IEEE Conference on Evolving and Adaptive Intelligent Systems
Volume	2017-May
ISSN (Print)	2330-4863
ISSN (Electronic)	2473-4691

Keywords

Empirical Data Analytics (EDA)
autonomous anomaly detection
data cloud
nonparametric

Access to Document

10.1109/EAIS.2017.7954831

Permanent link

Persistent link

Cite this

Gu, X., & Angelov, P. P. (2017). Autonomous anomaly detection. In I. Skrjanc, & S. Blazic (Eds.), IEEE Conference on Evolving and Adaptive Intelligent Systems 2017: Conference Proceedings Article 7954831 (IEEE Conference on Evolving and Adaptive Intelligent Systems; Vol. 2017-May). Institute of Electrical and Electronics Engineers. Advance online publication. https://doi.org/10.1109/EAIS.2017.7954831

@inproceedings{297c68b652244b33898383be75a631b9,

title = "Autonomous anomaly detection",

abstract = "In this paper, a new approach for autonomous anomaly detection is introduced within the Empirical Data Analytics (EDA) framework. This approach is fully data-driven and free from thresholds. Employing the nonparametric EDA estimators, the proposed approach can autonomously detect anomalies in an objective way based on the mutual distribution and ensemble properties of the data. The proposed approach firstly identifies the potential anomalies based on two EDA criteria, and then, partitions them into shape-free, non-parametric data clouds. Finally, it identifies the anomalies in regards to each data cloud (locally). Numerical examples based on synthetic and benchmark datasets demonstrate the validity and efficiency of the proposed approach.",

keywords = "Empirical Data Analytics (EDA), autonomous anomaly detection, data cloud, nonparametric",

author = "Xiaowei Gu and Angelov, \{Plamen Parvanov\}",

note = "Publisher Copyright: {\textcopyright} 2017 IEEE.",

year = "2017",

month = jun,

day = "22",

doi = "10.1109/EAIS.2017.7954831",

language = "English",

isbn = "9781509064441",

series = "IEEE Conference on Evolving and Adaptive Intelligent Systems",

publisher = "Institute of Electrical and Electronics Engineers",

editor = "Igor Skrjanc and Saso Blazic",

booktitle = "IEEE Conference on Evolving and Adaptive Intelligent Systems 2017",

address = "United States of America",

}

Autonomous anomaly detection. / Gu, Xiaowei; Angelov, Plamen Parvanov.
IEEE Conference on Evolving and Adaptive Intelligent Systems 2017: Conference Proceedings. ed. / Igor Skrjanc; Saso Blazic. Institute of Electrical and Electronics Engineers, 2017. 7954831 (IEEE Conference on Evolving and Adaptive Intelligent Systems; Vol. 2017-May).

Research output: Chapter in Book/Report/Conference proceeding › Conference Proceeding (ISBN)

TY - GEN

T1 - Autonomous anomaly detection

AU - Gu, Xiaowei

AU - Angelov, Plamen Parvanov

PY - 2017/6/22

Y1 - 2017/6/22

N2 - In this paper, a new approach for autonomous anomaly detection is introduced within the Empirical Data Analytics (EDA) framework. This approach is fully data-driven and free from thresholds. Employing the nonparametric EDA estimators, the proposed approach can autonomously detect anomalies in an objective way based on the mutual distribution and ensemble properties of the data. The proposed approach firstly identifies the potential anomalies based on two EDA criteria, and then, partitions them into shape-free, non-parametric data clouds. Finally, it identifies the anomalies in regards to each data cloud (locally). Numerical examples based on synthetic and benchmark datasets demonstrate the validity and efficiency of the proposed approach.

AB - In this paper, a new approach for autonomous anomaly detection is introduced within the Empirical Data Analytics (EDA) framework. This approach is fully data-driven and free from thresholds. Employing the nonparametric EDA estimators, the proposed approach can autonomously detect anomalies in an objective way based on the mutual distribution and ensemble properties of the data. The proposed approach firstly identifies the potential anomalies based on two EDA criteria, and then, partitions them into shape-free, non-parametric data clouds. Finally, it identifies the anomalies in regards to each data cloud (locally). Numerical examples based on synthetic and benchmark datasets demonstrate the validity and efficiency of the proposed approach.

KW - Empirical Data Analytics (EDA)

KW - autonomous anomaly detection

KW - data cloud

KW - nonparametric

UR - http://www.research.lancs.ac.uk/portal/en/publications/autonomous-anomaly-detection(358ecb1c-a5d8-4bad-929d-a74b0e43a871).html

UR - https://www.scopus.com/pages/publications/85054321162

U2 - 10.1109/EAIS.2017.7954831

DO - 10.1109/EAIS.2017.7954831

M3 - Conference Proceeding (ISBN)

SN - 9781509064441

T3 - IEEE Conference on Evolving and Adaptive Intelligent Systems

BT - IEEE Conference on Evolving and Adaptive Intelligent Systems 2017

A2 - Skrjanc, Igor

A2 - Blazic, Saso

PB - Institute of Electrical and Electronics Engineers

ER -

Autonomous anomaly detection

Abstract

Publication series

Keywords

Access to Document

Permanent link

Other files and links

Fingerprint

Cite this