Abstract
The current digital era of Industrial IoT and Automotive Technologies have made it standard for a large number of sensors to be installed on machines or vehicle, capture and exploit time-series data from such sensors for health monitoring tasks such as anomaly detection, fault detection, as well as prognostics. Anomalies or outliers are unexpected observations which deviate significantly from the expected observations and typically correspond to critical events. Current literature demonstrates good performance of Autoencoder for anomaly and novelty detection problems due to its efficient data encoding in an unsupervised manner. Despite the unsupervised nature of autoencoder-based anomaly detection methods, they are limited by the identification of anomalies using a threshold that is defined based on the distribution of reconstruction cost. Often, it is difficult to set a precise threshold when the distribution of reconstruction cost is not known. Motivated by this, we proposed a new unsupervised anomaly detection method (DeepConAD) that combined Autoencoder with forecasting model and used uncertainty estimates or confidence interval from forecasting model to identify anomalies in multivariate time series. We performed an experimental evaluation and comparison of DeepConAD with two other anomaly detection methods using Yahoo benchmark dataset, which contain both real and synthetic time series. Experimental results show DeepConAD outperforms other anomaly detection methods in most of the cases.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Wang, J.: Outlier detection in big data. J. Clean. Prod. 16(15), 2862 (2014)
Reunanen, N., Räty, T., Jokinen, J.J., Hoyt, T., Culler, D.: Unsupervised online detection and prediction of outliers in streams of sensor data. Int. J. Data Sci. Anal. 9(3), 285–314 (2019)
Da Xu, L., He, W., Li, S.: Internet of things in industries: a survey. IEEE Trans. Ind. Inform. 10(4), 2233–2243 (2014)
Teng, M.: Anomaly detection on time series. In: International Conference on Progress in Informatics and Computing, vol. 1, pp. 603–608 (2010)
Galeano, P., Pena, D., Tsay, R.S.: Outlier detection in multivariate time series by projection pursuit. J. Am. Stat. Assoc. 101, 654–669 (2006)
Gupta, M., Gao, J., Aggarwal, C.C., Han, J.: Outlier detection for temporal data: a survey. IEEE Trans. Knowl. Data Eng. 26(9), 2250–2267 (2014)
Theissler, A., Dear, I.: An anomaly detection approach to detect unexpected faults in recordings from test drives. In: Proceedings of the WASET International Conference on Vehicular Electronics and Safety 2013, vol. 7, no. 7, pp. 195–198 (2013)
Sangha, M.S., Yu, D.L., Gomm, J.B.: Sensor fault diagnosis for automotive engines with real data evaluation. Multicr. Int. J. Eng. Sci. Technol. 3(8), 13–25 (2011)
Kandhari, R., Chandola, V., Banerjee, A., Kumar, V., Kandhari, R.: Anomaly detection: a survey. ACM Comput. Surv. 41(3), 1–6 (2009)
Malhotra, P.A.P Vig, L., Shroff, G., Rinard, M.: Long short term memory networks for anomaly detection in time series. In: Proceedings, European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, Bruges (Belgium), 22–24 April 2015, (2015)
Aldosari, M.S.: Unsupervised anomaly detection in sequences using long short term memory recurrent neural networks. PhD Diss. Georg. Mason Univ., p. 98 (2016)
Saurav, S., Malhotra, P., Vishnu, T.V., Gugulothu, N., Vig, L., Agarwal, P., Shroff, G.: Online anomaly detection with concept drift adaptation using recurrent neural networks. In: Proceedings of the ACM India Joint International Conference on Data Science and Management of Data - CoDS-COMAD 2018, pp. 78–87 (2018)
Munir, M., Siddiqui, S.A., Dengel, A., Ahmed, S.: DeepAnT: a deep learning approach for unsupervised anomaly detection in time series. IEEE Access 7, 1991–2005 (2019)
Gugulothu, N., Vishnu, T.V., Malhotra, P., Vig, L., Agarwal, P., Shroo, G.: Predicting remaining useful life using time series embeddings based on recurrent neural networks. In: 2nd ML PHM Work. SIGKDD 2017, vol. 10 (2017)
Malhotra, P., Ramakrishnan, A., Anand, G., Vig, L., Agarwal, P., Shroff, G.: LSTM-based encoder-decoder for multi-sensor anomaly detection. arXiv.org, July 2016
Schreyer, M., Sattarov, T., Borth, D., Dengel, A., Reimer, B.: Detection of anomalies in large scale accounting data using deep autoencoder networks arXiv:1709.05254, September 2017
Amarbayasgalan, T., Jargalsaikhan, B., Ryu, K.: Unsupervised novelty detection using deep autoencoders with density based clustering. Appl. Sci. 8(9), 1468 (2018)
Zhu, L., Laptev, N.: Deep and confident prediction for time series at Uber. In: IEEE International Conference on Data Mining Workshops, ICDMW, vol. 2017–November, pp. 103–110 (2017)
Hyndman, R.J., Wang, E., Laptev, N.: Large-scale unusual time series detection. In: Proceedings - 15th IEEE International Conference on Data Mining Workshop, ICDMW 2015, pp. 1616–1619 (2016)
Keogh, E., Lonardi, S., Chiu, B.Y.: Finding surprising patterns in a time series database in linear time and space. In: Proceedings of the Eighth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD 2002, p. 550 (2002)
Keogh, E., Lin, J., Fu, A.: HOT SAX: efficiently finding the most unusual time series subsequence. In: Proceedings - IEEE International Conference on Data Mining, ICDM, pp. 226–233 (2005)
Bontemps, L., Cao, V.L., McDermott, J., Le-Khac, N.A.: Collective anomaly detection based on long short-term memory recurrent neural networks. In: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), LNCS, vol. 10018, pp. 141–152 (2016)
Chauhan, S., Vig, L.: Anomaly detection in ECG time signals via deep long short-term memory networks. In: Proceedings of the 2015 IEEE International Conference on Data Science and Advanced Analytics, DSAA 2015 (2015)
Hochreiter, S., Schmidhuber, J.J.: Long short-term memory. Neural Comput. 9(8), 1–32 (1997)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Tambuwal, A.I., Bello, A.M. (2020). DeepConAD: Deep and Confidence Prediction for Unsupervised Anomaly Detection in Time Series. In: Arai, K., Kapoor, S., Bhatia, R. (eds) Intelligent Computing. SAI 2020. Advances in Intelligent Systems and Computing, vol 1229. Springer, Cham. https://doi.org/10.1007/978-3-030-52246-9_16
Download citation
DOI: https://doi.org/10.1007/978-3-030-52246-9_16
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-52245-2
Online ISBN: 978-3-030-52246-9
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)