Model-centered Ensemble for Anomaly Detection in Time Series

Erick Trentini, Ticiana Coelho da Silva, Leopoldo Melo Junior, Jose F. de Macêdo

Abstract

Time-series anomalies detection is a fast-growing area of study, due to the exponential growth of new data produced by sensors in many different contexts as the Internet of Things (IOT). Many predictive models have been proposed, and they provide promising results in differentiating normal and anomalous points in a time-series. In this paper, we aim to find and combine the best models on detecting anomalous time series, so that their different strategies or parameters can contribute to the time series analysis. We propose TSPME-AD (stands for Time Series Prediction Model Ensemble for Anomaly Detection). TSPME-AD is a model-centered based ensemble that trains some of the state-of-the-art predictive models with different hyper-parameters and combines their anomaly scores with a weighted function. The efficacy of our proposal was demonstrated in two real-world time-series datasets, power demand, and electrocardiogram.

Download


Paper Citation


in Harvard Style

Trentini E., Coelho da Silva T., Melo Junior L. and F. de Macêdo J. (2020). Model-centered Ensemble for Anomaly Detection in Time Series.In Proceedings of the 12th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART, ISBN 978-989-758-395-7, pages 700-707. DOI: 10.5220/0008985507000707


in Bibtex Style

@conference{icaart20,
author={Erick Trentini and Ticiana Coelho da Silva and Leopoldo Melo Junior and Jose F. de Macêdo},
title={Model-centered Ensemble for Anomaly Detection in Time Series},
booktitle={Proceedings of the 12th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART,},
year={2020},
pages={700-707},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0008985507000707},
isbn={978-989-758-395-7},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 12th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART,
TI - Model-centered Ensemble for Anomaly Detection in Time Series
SN - 978-989-758-395-7
AU - Trentini E.
AU - Coelho da Silva T.
AU - Melo Junior L.
AU - F. de Macêdo J.
PY - 2020
SP - 700
EP - 707
DO - 10.5220/0008985507000707