Unsupervised Deep Learning for IoT Time Series

Abstract

IoT time series analysis has found numerous applications in a wide variety of areas, ranging from health informatics to network security. Nevertheless, the complex spatial temporal dynamics and high dimensionality of IoT time series make the analysis increasingly challenging. In recent years, the powerful feature extraction and representation learning capabilities of deep learning (DL) have provided an effective means for IoT time series analysis. However, few existing surveys on time series have systematically discussed unsupervised DL-based methods. To fill this void, we investigate unsupervised deep learning for IoT time series, i.e., unsupervised anomaly detection and clustering, under a unified framework. We also discuss the application scenarios, public datasets, existing challenges, and future research directions in this area.

22 pages, 8 figures, has been published by IEEE Internet of Things Journal

References in corpus (11)

• Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling
• Time-Series Anomaly Detection Service at Microsoft
• Anomaly Detection for a Water Treatment System Using Unsupervised Machine Learning
• Quasi-Recurrent Neural Networks
• Feature Encoding with AutoEncoders for Weakly-supervised Anomaly Detection
• Dilated Recurrent Neural Networks
• Time Series Anomaly Detection Using Convolutional Neural Networks and Transfer Learning
• Anomaly Detection in Multivariate Non-stationary Time Series for Automatic DBMS Diagnosis
• Clustering Time Series Data through Autoencoder-based Deep Learning Models
• Autoencoder-based time series clustering with energy applications
• TimeAutoML: Autonomous Representation Learning for Multivariate Irregularly Sampled Time Series