<abstract xmlns="http://www.w3.org/1999/xhtml"><p>Time-series are ordered sequences of discrete-time data. Due to their temporal dimension, anomaly detection techniques used in time-series have to take into consideration time correlations and other time-related particularities. Generally, in order to evaluate the quality of an anomaly detection technique, the confusion matrix and its derived metrics such as precision and recall are used. These metrics, however, do not take this temporal dimension into consideration. In this paper, we propose three metrics that can be used to evaluate the quality of a classification, while accounting for the temporal dimension found in time-series data.</p></abstract>

Time-series are ordered sequences of discrete-time data. Due to their temporal dimension, anomaly detection techniques used in time-series have to take into consideration time correlations and other time-related particularities. Generally, in order to evaluate the quality of an anomaly detection technique, the confusion matrix and its derived metrics such as precision and recall are used. These metrics, however, do not take this temporal dimension into consideration. In this paper, we propose three metrics that can be used to evaluate the quality of a classification, while accounting for the temporal dimension found in time-series data.

Evaluation metrics for anomaly detection algorithms in time-series

Acta Universitatis Sapientiae, Informatica

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.

{"article-title":"Evaluation metrics for anomaly detection algorithms in time-series"}

Time-series are ordered sequences of discrete-time data. Due to their temporal dimension, anomaly detection techniques used in time-series have to take into...

Evaluation metrics for anomaly detection algorithms in time-series

Published Online: Jan 21, 2020

Page range: 113 - 130

Received: Sep 17, 2019

DOI: https://doi.org/10.2478/ausi-2019-0008

Keywords
anomaly detection, classification, evaluation metrics

© 2019 György Kovács et al., published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.

Evaluation metrics for anomaly detection algorithms in time-series

Published Online: Jan 21, 2020

Page range: 113 - 130

Received: Sep 17, 2019

DOI: https://doi.org/10.2478/ausi-2019-0008

Keywordsanomaly detection, classification, evaluation metrics

© 2019 György Kovács et al., published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.

Keywords
anomaly detection, classification, evaluation metrics