Harbinger: A framework for integration and analysis of time series event detection methods

• Rebecca Salles Federal Center for Technological Education of Rio de Janeiro http://orcid.org/0000-0002-1001-3839
• Luciana Escobar Federal Center for Technological Education of Rio de Janeiro
• Lais Baroni Federal Center for Technological Education of Rio de Janeiro
• Roccio Zorrilla National Laboratory for Scientific Computing
• Artur Ziviani National Laboratory for Scientific Computing
• Vinicius Kreischer National Laboratory for Scientific Computing
• Flavia Delicato Fluminense Federal University
• Paulo F. Pires Federal Fluminense University
• Luciano Maia Petrobras
• Rafaelli Coutinho Federal Center for Technological Education of Rio de Janeiro
• Laura Assis Federal Center for Technological Education of Rio de Janeiro
• Eduardo Ogasawara Federal Center for Technological Education of Rio de Janeiro

DOI: https://doi.org/10.5753/sbbd.2020.13626

Abstract

When analyzing time series, it is possible to observe significant changes in the behavior of its observations that frequently characterize the occurrence of events. Events may appear as anomalies, change points, or frequent patterns. In literature, there are several methods for event detection. However, the search for a suitable method for a time series is not a simple task, especially considering that the nature of the events is often not known. In this context, this work presents Harbinger, a framework for integration and analysis of event detection methods. Harbinger was evaluated with synthetic and real data, where it was possible to verify that its functionalities foster the selection of methods and the understanding of detected events.

Keywords: time series, event detection, framework, anomalies, outliers, change points, detection methods, detection models

References

E. Calikus, S. Nowaczyk, A. Sant’Anna, and O. Dikmen. No free lunch but a cheaper supper: A general framework for streaming anomaly detection. Expert Systems with Applications, 155:113453, October 2020. ISSN 0957-4174.

R. Carmona. Statistical Analysis of Financial Data in R. Springer-Verlag New York, 2014.

H. Chen and N. Zhang. Graph-based change-point detection. The Annals of Statistics, 43(1):139–176, February 2015. ISSN 0090-5364.

D. De Paepe, S. V. Hautte, B. Steenwinckel, F. De Turck, F. Ongenae, O. Janssens, and S. V. Hoecke. A generalized matrix profile framework with support for contextual series analysis. Engineering Applications of Artificial Intelligence, 90, 2020. ISSN 0952-1976.

B. Eriksson, P. Barford, R. Bowden, N. Duffield, J. Sommers, and M. Roughan. Basisdetect: A model-based network event detection framework. In Proceedings of the 10th ACM SIGCOMM, page 451–464, New York, NY, USA, 2010. Association for Computing Machinery.

M. Gupta, J. Gao, C. Aggarwal, and J. Han. Outlier Detection for Temporal Data: A Survey. IEEE Transactions on Knowledge and Data Engineering, 2014. ISSN 1041-4347.

V. Guralnik and J. Srivastava. Event Detection from Time Series Data. In Proceedings of the Fifth ACM SIGKDD, KDD ’99, pages 33–42, New York, NY, USA, 1999. ACM. ISBN 978-1-58113-143-7.

G. Lu, Y. Zhou, C. Lu, and X. Li. A novel framework of change-point detection for machine monitoring. Mechanical Systems and Signal Processing, 83, 07 2016.

H. Raza, G. Prasad, and Y. Li. EWMA Model Based Shift-Detection Methods for Detecting Covariate Shifts in Non-Stationary Environments. Pattern Recogn., 48(3):659–669, March 2015. ISSN 0031-3203.

F. Rehbach, S. Moritz, S. Chandrasekaran, M. Rebolledo, M. Friese, and T. Bartz-Beielstein. GECCO 2018 Industrial Challenge: Monitoring of drinking-water quality. pages 1–7, 2018.

R. Salles, K. Belloze, F. Porto, P.H. Gonzalez, and E. Ogasawara. Nonstationary time series transformation methods: An experimental review. Knowledge-Based Systems, 164:274–291, 2019.

D. Silva, A. Simões, C. Cardoso, D. E. M. Oliveira, Y. Souto, L. E. G. Vignoli, R. Salles, H. S. C. Jr, A. Ziviani, E. Ogasawara, F. C. Delicato, P. F. Pires, H. L. C. P. Pinto, L. Maia, and F. Porto. A conceptual vision toward the management of machine learning models. In Proceedings of the ER Forum 2019, Salvador, Bahia, Brazil, volume 2469, pages 15–27, 2019.

J.-I. Takeuchi and K. Yamanishi. A unifying framework for detecting outliers and change points from time series. IEEE Transactions on Knowledge and Data Engineering, 18(4):482–492, 2006.

P. D. Talagala, R. J. Hyndman, K. Smith-Miles, S. Kandanaarachchi, and M. Muñoz. Anomaly Detection in Streaming Nonstationary Temporal Data. Journal of Computational and Graphical Statistics, 29(1):13–27, 2020.

Yahoo! Webscope. Labeled anomaly detection dataset. March 2015.

L. Xiong, C. Jiang, C. Xu, K. Yu, and S. Guo. A framework of change-point detection for multivariate hydrological series. Water Resources Research, 51, 09 2015.