Advanced search
1 file | 748.83 KB Add to list

Interpretable detection of unstable smart TV usage from power state logs

Author
Organization
Abstract
Power state logs from smart TVs are collected in order to construct a time-series representation of their usage. Time-series that belong to a TV exhibiting instability problems are classified accordingly. To do so, an automated feature extraction approach is used, together with linear classification methods in order to realise interpretable classification decisions. A normalized true positive rate of 0.84 ± 0.10 is obtained for the classification. The normalized true negative rate equals 0.80 ± 0.03. The final model returns a regularity statistic called the Approximate Entropy as its most important feature.
Keywords
user profiling, smart TV, time-series feature extraction, TSFRESH, logistic regression, approximate entropy

Downloads

  • paper 40.pdf
    • full text
    • |
    • open access
    • |
    • PDF
    • |
    • 748.83 KB

Citation

Please use this url to cite or link to this publication:

MLA
Mazaev, Tamir, et al. “Interpretable Detection of Unstable Smart TV Usage from Power State Logs.” Proceedings 2019, Advances in Data Mining, Applications and Theoretical Aspects, edited by Petra Perner, ibai publishing, 2019.
APA
Mazaev, T., Janssens, O., Van Gheel, D., Lanoye, L., Crevecoeur, G., & Van Hoecke, S. (2019). Interpretable detection of unstable smart TV usage from power state logs. In P. Perner (Ed.), Proceedings 2019, Advances in Data Mining, Applications and Theoretical Aspects. Leipzig: ibai publishing.
Chicago author-date
Mazaev, Tamir, Olivier Janssens, Dirk Van Gheel, Lieve Lanoye, Guillaume Crevecoeur, and Sofie Van Hoecke. 2019. “Interpretable Detection of Unstable Smart TV Usage from Power State Logs.” In Proceedings 2019, Advances in Data Mining, Applications and Theoretical Aspects, edited by Petra Perner. Leipzig: ibai publishing.
Chicago author-date (all authors)
Mazaev, Tamir, Olivier Janssens, Dirk Van Gheel, Lieve Lanoye, Guillaume Crevecoeur, and Sofie Van Hoecke. 2019. “Interpretable Detection of Unstable Smart TV Usage from Power State Logs.” In Proceedings 2019, Advances in Data Mining, Applications and Theoretical Aspects, ed by. Petra Perner. Leipzig: ibai publishing.
Vancouver
1.
Mazaev T, Janssens O, Van Gheel D, Lanoye L, Crevecoeur G, Van Hoecke S. Interpretable detection of unstable smart TV usage from power state logs. In: Perner P, editor. Proceedings 2019, Advances in Data Mining, Applications and Theoretical Aspects. Leipzig: ibai publishing; 2019.
IEEE
[1]
T. Mazaev, O. Janssens, D. Van Gheel, L. Lanoye, G. Crevecoeur, and S. Van Hoecke, “Interpretable detection of unstable smart TV usage from power state logs,” in Proceedings 2019, Advances in Data Mining, Applications and Theoretical Aspects, New York, 2019.
@inproceedings{8633813,
  abstract     = {Power state logs from smart TVs are collected in order to construct a time-series representation of their usage. Time-series that belong to a TV exhibiting instability problems are classified accordingly. To do so, an automated feature extraction approach is used, together with linear classification methods in order to realise interpretable classification decisions. A normalized true positive rate of 0.84 ± 0.10 is obtained for the classification. The normalized true negative rate equals 0.80 ± 0.03. The final model returns a regularity statistic called the Approximate Entropy as its most important feature.},
  author       = {Mazaev, Tamir and Janssens, Olivier and Van Gheel, Dirk and Lanoye, Lieve and Crevecoeur, Guillaume and Van Hoecke, Sofie},
  booktitle    = {Proceedings 2019, Advances in Data Mining, Applications and Theoretical Aspects},
  editor       = {Perner, Petra},
  isbn         = {978-3-942952-60-6},
  keywords     = {user profiling,smart TV,time-series feature extraction,TSFRESH,logistic regression,approximate entropy},
  language     = {eng},
  location     = {New York},
  publisher    = {ibai publishing},
  title        = {Interpretable detection of unstable smart TV usage from power state logs},
  year         = {2019},
}