Anomaly detection based on data stream monitoring and prediction with improved Gaussian process regression algorithm

Jingyue Pang, Datong Liu, Haitao Liao, Yu Peng, Xiyuan Peng

Research output: Chapter in Book/Report/Conference proceedingConference contribution

17 Citations (Scopus)

Abstract

Condition monitoring has gradually become the necessary part of the diagnostics and prognostics for the complex systems. Especially, with the rapid development of data acquisition and communication technology, the appearing of large scale data set and data stream brings great challenges to model and process the condition monitoring data As a result, anomaly detection of the streaming monitoring data attracts more attention in the fields of prognostics and health management (PHM). Hence, in this study, Gaussian process regression (GPR) is applied for the abnormal detection in data stream; and on this basis a real-time abnormal detection method is proposed based on the improved anomaly detection and mitigation (IADAM) strategy and GPR which realizes incremental detecting for future data samples and requires no pre-classification labels of anomalies. Anomaly detection tested on an artificial data set and actual mobile traffic data set indicates the effectiveness and reasonability of IADAM-GPR model compared with naïve and Multilayer Perceptron (MLP) models.

Original languageEnglish (US)
Title of host publication2014 International Conference on Prognostics and Health Management, PHM 2014
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Print)9781479959426
DOIs
StatePublished - Feb 9 2015
Event2014 International Conference on Prognostics and Health Management, PHM 2014 - Cheney, United States
Duration: Jun 22 2014Jun 25 2014

Other

Other2014 International Conference on Prognostics and Health Management, PHM 2014
CountryUnited States
CityCheney
Period6/22/146/25/14

Fingerprint

Condition monitoring
Monitoring
Neural Networks (Computer)
Multilayer neural networks
Large scale systems
Labels
Data acquisition
Health
Technology
Communication
Datasets

Keywords

  • Anomaly detection and mitigation
  • Anomoly deteciton
  • Data stream
  • Gaussian process regression
  • Hypothesis testing

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Computer Science Applications
  • Software
  • Health Information Management

Cite this

Pang, J., Liu, D., Liao, H., Peng, Y., & Peng, X. (2015). Anomaly detection based on data stream monitoring and prediction with improved Gaussian process regression algorithm. In 2014 International Conference on Prognostics and Health Management, PHM 2014 [7036394] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICPHM.2014.7036394

Anomaly detection based on data stream monitoring and prediction with improved Gaussian process regression algorithm. / Pang, Jingyue; Liu, Datong; Liao, Haitao; Peng, Yu; Peng, Xiyuan.

2014 International Conference on Prognostics and Health Management, PHM 2014. Institute of Electrical and Electronics Engineers Inc., 2015. 7036394.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Pang, J, Liu, D, Liao, H, Peng, Y & Peng, X 2015, Anomaly detection based on data stream monitoring and prediction with improved Gaussian process regression algorithm. in 2014 International Conference on Prognostics and Health Management, PHM 2014., 7036394, Institute of Electrical and Electronics Engineers Inc., 2014 International Conference on Prognostics and Health Management, PHM 2014, Cheney, United States, 6/22/14. https://doi.org/10.1109/ICPHM.2014.7036394
Pang J, Liu D, Liao H, Peng Y, Peng X. Anomaly detection based on data stream monitoring and prediction with improved Gaussian process regression algorithm. In 2014 International Conference on Prognostics and Health Management, PHM 2014. Institute of Electrical and Electronics Engineers Inc. 2015. 7036394 https://doi.org/10.1109/ICPHM.2014.7036394
Pang, Jingyue ; Liu, Datong ; Liao, Haitao ; Peng, Yu ; Peng, Xiyuan. / Anomaly detection based on data stream monitoring and prediction with improved Gaussian process regression algorithm. 2014 International Conference on Prognostics and Health Management, PHM 2014. Institute of Electrical and Electronics Engineers Inc., 2015.
@inproceedings{e3fe2936fad647ba8e56f57f4ad9f4e8,
title = "Anomaly detection based on data stream monitoring and prediction with improved Gaussian process regression algorithm",
abstract = "Condition monitoring has gradually become the necessary part of the diagnostics and prognostics for the complex systems. Especially, with the rapid development of data acquisition and communication technology, the appearing of large scale data set and data stream brings great challenges to model and process the condition monitoring data As a result, anomaly detection of the streaming monitoring data attracts more attention in the fields of prognostics and health management (PHM). Hence, in this study, Gaussian process regression (GPR) is applied for the abnormal detection in data stream; and on this basis a real-time abnormal detection method is proposed based on the improved anomaly detection and mitigation (IADAM) strategy and GPR which realizes incremental detecting for future data samples and requires no pre-classification labels of anomalies. Anomaly detection tested on an artificial data set and actual mobile traffic data set indicates the effectiveness and reasonability of IADAM-GPR model compared with na{\"i}ve and Multilayer Perceptron (MLP) models.",
keywords = "Anomaly detection and mitigation, Anomoly deteciton, Data stream, Gaussian process regression, Hypothesis testing",
author = "Jingyue Pang and Datong Liu and Haitao Liao and Yu Peng and Xiyuan Peng",
year = "2015",
month = "2",
day = "9",
doi = "10.1109/ICPHM.2014.7036394",
language = "English (US)",
isbn = "9781479959426",
booktitle = "2014 International Conference on Prognostics and Health Management, PHM 2014",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Anomaly detection based on data stream monitoring and prediction with improved Gaussian process regression algorithm

AU - Pang, Jingyue

AU - Liu, Datong

AU - Liao, Haitao

AU - Peng, Yu

AU - Peng, Xiyuan

PY - 2015/2/9

Y1 - 2015/2/9

N2 - Condition monitoring has gradually become the necessary part of the diagnostics and prognostics for the complex systems. Especially, with the rapid development of data acquisition and communication technology, the appearing of large scale data set and data stream brings great challenges to model and process the condition monitoring data As a result, anomaly detection of the streaming monitoring data attracts more attention in the fields of prognostics and health management (PHM). Hence, in this study, Gaussian process regression (GPR) is applied for the abnormal detection in data stream; and on this basis a real-time abnormal detection method is proposed based on the improved anomaly detection and mitigation (IADAM) strategy and GPR which realizes incremental detecting for future data samples and requires no pre-classification labels of anomalies. Anomaly detection tested on an artificial data set and actual mobile traffic data set indicates the effectiveness and reasonability of IADAM-GPR model compared with naïve and Multilayer Perceptron (MLP) models.

AB - Condition monitoring has gradually become the necessary part of the diagnostics and prognostics for the complex systems. Especially, with the rapid development of data acquisition and communication technology, the appearing of large scale data set and data stream brings great challenges to model and process the condition monitoring data As a result, anomaly detection of the streaming monitoring data attracts more attention in the fields of prognostics and health management (PHM). Hence, in this study, Gaussian process regression (GPR) is applied for the abnormal detection in data stream; and on this basis a real-time abnormal detection method is proposed based on the improved anomaly detection and mitigation (IADAM) strategy and GPR which realizes incremental detecting for future data samples and requires no pre-classification labels of anomalies. Anomaly detection tested on an artificial data set and actual mobile traffic data set indicates the effectiveness and reasonability of IADAM-GPR model compared with naïve and Multilayer Perceptron (MLP) models.

KW - Anomaly detection and mitigation

KW - Anomoly deteciton

KW - Data stream

KW - Gaussian process regression

KW - Hypothesis testing

UR - http://www.scopus.com/inward/record.url?scp=84929587516&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84929587516&partnerID=8YFLogxK

U2 - 10.1109/ICPHM.2014.7036394

DO - 10.1109/ICPHM.2014.7036394

M3 - Conference contribution

AN - SCOPUS:84929587516

SN - 9781479959426

BT - 2014 International Conference on Prognostics and Health Management, PHM 2014

PB - Institute of Electrical and Electronics Engineers Inc.

ER -