Unsupervised anomaly detection with LSTM autoencoders using statistical data-filtering

• Published in: Applied soft computing Vol. 108; p. 107443
• Main Authors: Maleki, Sepehr, Maleki, Sasan, Jennings, Nicholas R.
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.09.2021
• Subjects: Anomaly detection; Autoencoders; Changepoint detection; IoT; LSTM networks; Changepoint detection; Autoencoders; LSTM networks; Anomaly detection; IoT

To address one of the most challenging industry problems, we develop an enhanced training algorithm for anomaly detection in unlabelled sequential data such as time-series. We propose the outputs of a well-designed system are drawn from an unknown probability distribution, U, in normal conditions. W...