Denoising Architecture for Unsupervised Anomaly Detection in Time-Series

Skaf, Wadie; Horváth, Tomáš

Abstract

Anomalies in time-series provide insights of critical scenarios across a range of industries, from banking and aerospace to information technology, security, and medicine. However, identifying anomalies in time-series data is particularly challenging due to the imprecise definition of anomalies, the frequent absence of labels, and the enormously complex temporal correlations present in such data. The LSTM Autoencoder is an Encoder-Decoder scheme for Anomaly Detection based on Long Short Term Memory Networks that learns to reconstruct time-series behavior and then uses reconstruction error to identify abnormalities. We introduce the Denoising Architecture as a complement to this LSTM Encoder-Decoder model and investigate its effect on real-world as well as artificially generated datasets. We demonstrate that the proposed architecture increases both the accuracy and the training speed, thereby, making the LSTM Autoencoder more efficient for unsupervised anomaly detection tasks.

Citation

Skaf, W., & Horváth, T. (2022, September). Denoising Architecture for Unsupervised Anomaly Detection in Time-Series. Presented at ADBIS 2022: 26th European Conference on Advances in Databases and Information Systems, Turin, Italy

• Presentation Conference Type: Conference Paper (published)
• Conference Name: ADBIS 2022: 26th European Conference on Advances in Databases and Information Systems
• Start Date: Sep 5, 2022
• End Date: Sep 8, 2022
• Online Publication Date: Aug 29, 2022
• Publication Date: 2022
• Deposit Date: Apr 8, 2024
• Publisher: Springer
• Pages: 178-187
• Series Title: Communications in Computer and Information Science
• Series Number: 1652
• Series ISSN: 1865-0929
• Book Title: New Trends in Database and Information Systems: ADBIS 2022 Short Papers, Doctoral Consortium and Workshops: DOING, K-GALS, MADEISD, MegaData, SWODCH, Turin, Italy, September 5–8, 2022, Proceedings
• ISBN: 9783031157424
• DOI: https://doi.org/10.1007/978-3-031-15743-1_17
• Keywords: Anomaly detection, Time-series, Autoencoder
• Public URL: http://researchrepository.napier.ac.uk/Output/3587437