An efficient method for anomaly detection in non-stationary data streams

Milad Chenaghlou, Masud Moshtaghi, Christopher Leckie, Mahsa Salehi

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

    Abstract

    Anomaly detection in data streams has become a major research problem in the era of ubiquitous sensing. We are collecting large amounts of data from non-stationary environments, which makes traditional anomaly detection techniques ineffective. In this paper we propose an unsupervised cluster-based algorithm for modelling normal behaviour in nonstationary data streams and detecting anomalous data points. We show that our method scales linearly with the number of observed data points, while the complexity of our model is independent of the size of the data stream. We have employed a selective clustering approach to optimize the computation time needed to model the normal data. Our experiments on largescale synthetic and real life datasets show that the accuracy of the proposed algorithm is comparable to the state-of-the-art techniques reported in the literature while providing substantial improvements in terms of computation time.

    Original languageEnglish
    Title of host publication2017 IEEE Global Communications Conference (GLOBECOM) - Proceedings
    Subtitle of host publicationSingapore, 4-8 December 2017
    EditorsYing-Chang Liang, Teng Joon Lim, Chengshan Xiao
    Place of PublicationPiscataway NJ USA
    PublisherIEEE, Institute of Electrical and Electronics Engineers
    Pages1-6
    Number of pages6
    ISBN (Electronic)9781509050185
    ISBN (Print)9781509050192
    DOIs
    Publication statusPublished - 2017
    EventIEEE Global Telecommunications Conference 2017 - Marina Bay Sands Expo and Convention Centre, Singapore, Singapore
    Duration: 4 Dec 20178 Dec 2017
    https://globecom2017.ieee-globecom.org/

    Conference

    ConferenceIEEE Global Telecommunications Conference 2017
    Abbreviated titleGLOBECOM 2017
    CountrySingapore
    CitySingapore
    Period4/12/178/12/17
    Internet address

    Cite this

    Chenaghlou, M., Moshtaghi, M., Leckie, C., & Salehi, M. (2017). An efficient method for anomaly detection in non-stationary data streams. In Y-C. Liang, T. J. Lim, & C. Xiao (Eds.), 2017 IEEE Global Communications Conference (GLOBECOM) - Proceedings: Singapore, 4-8 December 2017 (pp. 1-6). Piscataway NJ USA: IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/GLOCOM.2017.8255032
    Chenaghlou, Milad ; Moshtaghi, Masud ; Leckie, Christopher ; Salehi, Mahsa. / An efficient method for anomaly detection in non-stationary data streams. 2017 IEEE Global Communications Conference (GLOBECOM) - Proceedings: Singapore, 4-8 December 2017. editor / Ying-Chang Liang ; Teng Joon Lim ; Chengshan Xiao. Piscataway NJ USA : IEEE, Institute of Electrical and Electronics Engineers, 2017. pp. 1-6
    @inproceedings{e7c46c8210f9411e8dbae3d0fd9f9d93,
    title = "An efficient method for anomaly detection in non-stationary data streams",
    abstract = "Anomaly detection in data streams has become a major research problem in the era of ubiquitous sensing. We are collecting large amounts of data from non-stationary environments, which makes traditional anomaly detection techniques ineffective. In this paper we propose an unsupervised cluster-based algorithm for modelling normal behaviour in nonstationary data streams and detecting anomalous data points. We show that our method scales linearly with the number of observed data points, while the complexity of our model is independent of the size of the data stream. We have employed a selective clustering approach to optimize the computation time needed to model the normal data. Our experiments on largescale synthetic and real life datasets show that the accuracy of the proposed algorithm is comparable to the state-of-the-art techniques reported in the literature while providing substantial improvements in terms of computation time.",
    author = "Milad Chenaghlou and Masud Moshtaghi and Christopher Leckie and Mahsa Salehi",
    year = "2017",
    doi = "10.1109/GLOCOM.2017.8255032",
    language = "English",
    isbn = "9781509050192",
    pages = "1--6",
    editor = "Ying-Chang Liang and Lim, {Teng Joon} and Chengshan Xiao",
    booktitle = "2017 IEEE Global Communications Conference (GLOBECOM) - Proceedings",
    publisher = "IEEE, Institute of Electrical and Electronics Engineers",
    address = "United States of America",

    }

    Chenaghlou, M, Moshtaghi, M, Leckie, C & Salehi, M 2017, An efficient method for anomaly detection in non-stationary data streams. in Y-C Liang, TJ Lim & C Xiao (eds), 2017 IEEE Global Communications Conference (GLOBECOM) - Proceedings: Singapore, 4-8 December 2017. IEEE, Institute of Electrical and Electronics Engineers, Piscataway NJ USA, pp. 1-6, IEEE Global Telecommunications Conference 2017, Singapore, Singapore, 4/12/17. https://doi.org/10.1109/GLOCOM.2017.8255032

    An efficient method for anomaly detection in non-stationary data streams. / Chenaghlou, Milad; Moshtaghi, Masud; Leckie, Christopher; Salehi, Mahsa.

    2017 IEEE Global Communications Conference (GLOBECOM) - Proceedings: Singapore, 4-8 December 2017. ed. / Ying-Chang Liang; Teng Joon Lim; Chengshan Xiao. Piscataway NJ USA : IEEE, Institute of Electrical and Electronics Engineers, 2017. p. 1-6.

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

    TY - GEN

    T1 - An efficient method for anomaly detection in non-stationary data streams

    AU - Chenaghlou, Milad

    AU - Moshtaghi, Masud

    AU - Leckie, Christopher

    AU - Salehi, Mahsa

    PY - 2017

    Y1 - 2017

    N2 - Anomaly detection in data streams has become a major research problem in the era of ubiquitous sensing. We are collecting large amounts of data from non-stationary environments, which makes traditional anomaly detection techniques ineffective. In this paper we propose an unsupervised cluster-based algorithm for modelling normal behaviour in nonstationary data streams and detecting anomalous data points. We show that our method scales linearly with the number of observed data points, while the complexity of our model is independent of the size of the data stream. We have employed a selective clustering approach to optimize the computation time needed to model the normal data. Our experiments on largescale synthetic and real life datasets show that the accuracy of the proposed algorithm is comparable to the state-of-the-art techniques reported in the literature while providing substantial improvements in terms of computation time.

    AB - Anomaly detection in data streams has become a major research problem in the era of ubiquitous sensing. We are collecting large amounts of data from non-stationary environments, which makes traditional anomaly detection techniques ineffective. In this paper we propose an unsupervised cluster-based algorithm for modelling normal behaviour in nonstationary data streams and detecting anomalous data points. We show that our method scales linearly with the number of observed data points, while the complexity of our model is independent of the size of the data stream. We have employed a selective clustering approach to optimize the computation time needed to model the normal data. Our experiments on largescale synthetic and real life datasets show that the accuracy of the proposed algorithm is comparable to the state-of-the-art techniques reported in the literature while providing substantial improvements in terms of computation time.

    UR - http://www.scopus.com/inward/record.url?scp=85046419416&partnerID=8YFLogxK

    U2 - 10.1109/GLOCOM.2017.8255032

    DO - 10.1109/GLOCOM.2017.8255032

    M3 - Conference Paper

    SN - 9781509050192

    SP - 1

    EP - 6

    BT - 2017 IEEE Global Communications Conference (GLOBECOM) - Proceedings

    A2 - Liang, Ying-Chang

    A2 - Lim, Teng Joon

    A2 - Xiao, Chengshan

    PB - IEEE, Institute of Electrical and Electronics Engineers

    CY - Piscataway NJ USA

    ER -

    Chenaghlou M, Moshtaghi M, Leckie C, Salehi M. An efficient method for anomaly detection in non-stationary data streams. In Liang Y-C, Lim TJ, Xiao C, editors, 2017 IEEE Global Communications Conference (GLOBECOM) - Proceedings: Singapore, 4-8 December 2017. Piscataway NJ USA: IEEE, Institute of Electrical and Electronics Engineers. 2017. p. 1-6 https://doi.org/10.1109/GLOCOM.2017.8255032