A support vector machine based technique for online detection of outliers in transient time series

Hugo Martins, Luis Palma, Alberto Cardoso, Paulo Gil

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

7 Citations (Scopus)

Abstract

This paper deals with online detection and accommodation of outliers in transient time series by appealing to a machine learning technique. The methodology is based on a Least Squares Support Vector Machine technique together with a sliding window-based learning algorithm. A modification to this method is proposed so as to extend its application to transient raw data collected from transmitters attached to a Wireless Sensor Network. The performance of two approaches are compared on a particular controlled data set.
Original languageEnglish
Title of host publication2015 10th Asian Control Conference (ASCC)
Subtitle of host publicationEmerging Control Techniques for a Sustainable World
EditorsH. Selamat , H. R. H. Ramli , A. A. M. Faudzi , R. Z. A. Rahman , A. J. Ishak , A. C. Soh , S. A. Ahmad
Place of PublicationNew York
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
ISBN (Electronic)978-1-4799-7862-5
ISBN (Print)978-1-4799-7863-2
DOIs
Publication statusPublished - 2015
Event10th Asian Control Conference (ASCC 2015) - Kota Kinabalu, Malaysia
Duration: 31 May 20153 Jun 2015
Conference number: 10th

Conference

Conference10th Asian Control Conference (ASCC 2015)
Abbreviated titleASCC 2015
CountryMalaysia
CityKota Kinabalu
Period31/05/153/06/15

Fingerprint

Support vector machines
Time series
Learning algorithms
Learning systems
Wireless sensor networks
Transmitters

Keywords

  • WIRELESS SENSOR NETWORKS
  • ANOMALY DETECTION

Cite this

Martins, H., Palma, L., Cardoso, A., & Gil, P. (2015). A support vector machine based technique for online detection of outliers in transient time series. In H. Selamat , H. R. H. Ramli , A. A. M. Faudzi , R. Z. A. Rahman , A. J. Ishak , A. C. Soh , & S. A. Ahmad (Eds.), 2015 10th Asian Control Conference (ASCC): Emerging Control Techniques for a Sustainable World [7244794] New York: Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.1109/ascc.2015.7244794
Martins, Hugo ; Palma, Luis ; Cardoso, Alberto ; Gil, Paulo. / A support vector machine based technique for online detection of outliers in transient time series. 2015 10th Asian Control Conference (ASCC): Emerging Control Techniques for a Sustainable World. editor / H. Selamat ; H. R. H. Ramli ; A. A. M. Faudzi ; R. Z. A. Rahman ; A. J. Ishak ; A. C. Soh ; S. A. Ahmad . New York : Institute of Electrical and Electronics Engineers (IEEE), 2015.
@inproceedings{972077c4b7af4ee5a25d11f88bfd65b6,
title = "A support vector machine based technique for online detection of outliers in transient time series",
abstract = "This paper deals with online detection and accommodation of outliers in transient time series by appealing to a machine learning technique. The methodology is based on a Least Squares Support Vector Machine technique together with a sliding window-based learning algorithm. A modification to this method is proposed so as to extend its application to transient raw data collected from transmitters attached to a Wireless Sensor Network. The performance of two approaches are compared on a particular controlled data set.",
keywords = "WIRELESS SENSOR NETWORKS , ANOMALY DETECTION",
author = "Hugo Martins and Luis Palma and Alberto Cardoso and Paulo Gil",
note = "Sem PDF conforme Despacho. This work has been partially supported by Project CENTRO-07-ST24-FEDER-002003 (iCIS-Intelligent Computing in the Internet of Services).",
year = "2015",
doi = "10.1109/ascc.2015.7244794",
language = "English",
isbn = "978-1-4799-7863-2",
editor = "{Selamat }, H. and {Ramli }, {H. R. H.} and {Faudzi }, {A. A. M.} and {Rahman }, {R. Z. A.} and {Ishak }, {A. J.} and {Soh }, {A. C.} and {Ahmad }, {S. A.}",
booktitle = "2015 10th Asian Control Conference (ASCC)",
publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

}

Martins, H, Palma, L, Cardoso, A & Gil, P 2015, A support vector machine based technique for online detection of outliers in transient time series. in H Selamat , HRH Ramli , AAM Faudzi , RZA Rahman , AJ Ishak , AC Soh & SA Ahmad (eds), 2015 10th Asian Control Conference (ASCC): Emerging Control Techniques for a Sustainable World., 7244794, Institute of Electrical and Electronics Engineers (IEEE), New York, 10th Asian Control Conference (ASCC 2015), Kota Kinabalu, Malaysia, 31/05/15. https://doi.org/10.1109/ascc.2015.7244794

A support vector machine based technique for online detection of outliers in transient time series. / Martins, Hugo; Palma, Luis; Cardoso, Alberto; Gil, Paulo.

2015 10th Asian Control Conference (ASCC): Emerging Control Techniques for a Sustainable World. ed. / H. Selamat ; H. R. H. Ramli ; A. A. M. Faudzi ; R. Z. A. Rahman ; A. J. Ishak ; A. C. Soh ; S. A. Ahmad . New York : Institute of Electrical and Electronics Engineers (IEEE), 2015. 7244794.

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

TY - GEN

T1 - A support vector machine based technique for online detection of outliers in transient time series

AU - Martins, Hugo

AU - Palma, Luis

AU - Cardoso, Alberto

AU - Gil, Paulo

N1 - Sem PDF conforme Despacho. This work has been partially supported by Project CENTRO-07-ST24-FEDER-002003 (iCIS-Intelligent Computing in the Internet of Services).

PY - 2015

Y1 - 2015

N2 - This paper deals with online detection and accommodation of outliers in transient time series by appealing to a machine learning technique. The methodology is based on a Least Squares Support Vector Machine technique together with a sliding window-based learning algorithm. A modification to this method is proposed so as to extend its application to transient raw data collected from transmitters attached to a Wireless Sensor Network. The performance of two approaches are compared on a particular controlled data set.

AB - This paper deals with online detection and accommodation of outliers in transient time series by appealing to a machine learning technique. The methodology is based on a Least Squares Support Vector Machine technique together with a sliding window-based learning algorithm. A modification to this method is proposed so as to extend its application to transient raw data collected from transmitters attached to a Wireless Sensor Network. The performance of two approaches are compared on a particular controlled data set.

KW - WIRELESS SENSOR NETWORKS

KW - ANOMALY DETECTION

U2 - 10.1109/ascc.2015.7244794

DO - 10.1109/ascc.2015.7244794

M3 - Conference contribution

SN - 978-1-4799-7863-2

BT - 2015 10th Asian Control Conference (ASCC)

A2 - Selamat , H.

A2 - Ramli , H. R. H.

A2 - Faudzi , A. A. M.

A2 - Rahman , R. Z. A.

A2 - Ishak , A. J.

A2 - Soh , A. C.

A2 - Ahmad , S. A.

PB - Institute of Electrical and Electronics Engineers (IEEE)

CY - New York

ER -

Martins H, Palma L, Cardoso A, Gil P. A support vector machine based technique for online detection of outliers in transient time series. In Selamat H, Ramli HRH, Faudzi AAM, Rahman RZA, Ishak AJ, Soh AC, Ahmad SA, editors, 2015 10th Asian Control Conference (ASCC): Emerging Control Techniques for a Sustainable World. New York: Institute of Electrical and Electronics Engineers (IEEE). 2015. 7244794 https://doi.org/10.1109/ascc.2015.7244794