Team for Research in
Ubiquitous Secure Technology

Citation
Bonnie Zhu, Shankar Sastry. "Revisit Dynamic ARIMA Based Anomaly Detection". SocialCom/PASSAT 2011, IEEE, pp.1263-1268, 9, October, 2011.

Abstract
On the assumption that a model is correctly learned and built, the typical usage of ARIMA in anomaly detection compares data points with those predicated through the model to determine whether anomalies occur. Yet the time variability by the coefficients in those dynamic regression models is possibly indicative of whether anomalies are in the data set on which the ARIMA model builds. Thus we introduce a corresponding framework and a novel anomaly detection method that combines the Kalman filter for identifying the parameters of those dynamic models with a General Likelihood Ratio (GLR) test that is based on the former for detecting suspicious changes in the parameters and therefore the models. We illustrate the idea through experiments and show its promising potential in terms of accuracy and robustness.

Electronic downloads

• http://dx.doi.org/10.1109/PASSAT/SocialCom.2011.84

• HTML

  Bonnie Zhu, Shankar Sastry. <a
  href="http://www.truststc.org/pubs/860.html"
  >Revisit Dynamic ARIMA Based Anomaly Detection</a>,
  SocialCom/PASSAT 2011, IEEE, pp.1263-1268, 9, October, 2011.

• Plain text

  Bonnie Zhu, Shankar Sastry. "Revisit Dynamic ARIMA
  Based Anomaly Detection". SocialCom/PASSAT 2011, IEEE,
  pp.1263-1268, 9, October, 2011.

• BibTeX

  @inproceedings{ZhuSastry11_RevisitDynamicARIMABasedAnomalyDetection,
      author = {Bonnie Zhu and Shankar Sastry},
      title = {Revisit Dynamic ARIMA Based Anomaly Detection},
      booktitle = {SocialCom/PASSAT 2011},
      organization = {IEEE},
      pages = {pp.1263-1268},
      day = {9},
      month = {October},
      year = {2011},
      abstract = {On the assumption that a model is correctly
                learned and built, the typical usage of ARIMA in
                anomaly detection compares data points with those
                predicated through the model to determine whether
                anomalies occur. Yet the time variability by the
                coefficients in those dynamic regression models is
                possibly indicative of whether anomalies are in
                the data set on which the ARIMA model builds. Thus
                we introduce a corresponding framework and a novel
                anomaly detection method that combines the Kalman
                filter for identifying the parameters of those
                dynamic models with a General Likelihood Ratio
                (GLR) test that is based on the former for
                detecting suspicious changes in the parameters and
                therefore the models. We illustrate the idea
                through experiments and show its promising
                potential in terms of accuracy and robustness.},
      URL = {http://www.truststc.org/pubs/860.html}
  }
  

Posted by Mary Stewart on 4 Apr 2012.
For additional information, see the Publications FAQ or contact webmaster at www truststc org.