TGAN-AD: Transformer-Based GAN for Anomaly Detection of Time Series Data

Article Chemistry, Multidisciplinary

Dual Auto-Encoder GAN-Based Anomaly Detection for Industrial Control System

Lei Chen et al.

Summary: This study proposes a dual auto-encoder GAN-based anomaly detection model, called DAGAN, for industrial control systems. The model can accurately and efficiently detect anomalies without any abnormal samples, and utilizes a dynamic strategy to learn the marginal distribution of the training data, reducing the probability of misjudging marginal samples.

APPLIED SCIENCES-BASEL (2022)

Add to Collection

Article Chemistry, Multidisciplinary

Outlier Detection for Multivariate Time Series Using Dynamic Bayesian Networks

Jorge L. Serras et al.

Summary: This study proposes a complete and automatic outlier detection system using a dynamic Bayesian network modeling algorithm, which is validated on simulated and real data to demonstrate its effectiveness in capturing anomalies in time series data.

APPLIED SCIENCES-BASEL (2021)

Add to Collection

Article Chemistry, Multidisciplinary

Time Series Anomaly Detection for KPIs Based on Correlation Analysis and HMM

Zijing Shang et al.

Summary: This paper presents a time series anomaly detection method based on correlation analysis and HMM, which detects anomalies in KPIs in distributed systems by analyzing residual sequences and constructing correlation matrices. The experiments on public data sets show that the method achieves good results.

APPLIED SCIENCES-BASEL (2021)

Add to Collection

Article Chemistry, Multidisciplinary

Semi-Supervised Time Series Anomaly Detection Based on Statistics and Deep Learning

Jehn-Ruey Jiang et al.

Summary: This paper proposed a semi-supervised anomaly detection framework, Tri-CAD, for univariate time series, and experiments showed that it outperformed other related methods in terms of precision, recall, and F-1-score.

APPLIED SCIENCES-BASEL (2021)

Add to Collection

Article Chemistry, Multidisciplinary