Unsupervised Anomaly Detection in Time Series Using LSTM-Based Autoencoders

Provotar, O. I.; Linder, Yaroslav M.; Veres, Maksym M.

doi:10.1109/atit49449.2019.9030505

Cited by 99 publications

(50 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This has a practical meaning since anomalous data are not always available or it is impossible to cover all the types of these data. Many advantages of using the autoencoder approach have been discussed in (Provotar et al, 2019). The use of LSTM autoencoder for anomaly detection on multivariate time series data can be seen in several studies, for example, Pereira and Silveira (2018) and Principi et al (2019).…”

Section: The Needed Conceptsmentioning

confidence: 99%

Forecasting and Anomaly Detection approaches using LSTM and LSTM Autoencoder techniques with the applications in supply chain management

Nguyen

Tran

Thomassey

et al. 2021

International Journal of Information Management

323

View full text Add to dashboard Cite

Making appropriate decisions is indeed a key factor to help companies facing challenges from supply chains nowadays. In this paper, we propose two datadriven approaches that allow making better decisions in supply chain management. In particular, we suggest a Long Short Term Memory (LSTM) networkbased method for forecasting multivariate time series data and an LSTM Autoencoder network-based method combined with a one-class support vector machine algorithm for detecting anomalies in sales. Unlike other approaches, we recommend combining external and internal company data sources for the purpose of enhancing the performance of forecasting algorithms using multivariate LSTM with the optimal hyperparameters. In addition, we also propose a method to optimize hyperparameters for hybrid algorithms for detecting anomalies in time series data. The proposed approaches will be applied to both benchmarking datasets and real data in fashion retail. The obtained results show that the LSTM Autoencoder based method leads to better performance for anomaly detection compared to the LSTM based method suggested in a previous study.The proposed forecasting method for multivariate time series data also performs better some other methods based on a dataset provided by NASA.

show abstract

Section: The Needed Conceptsmentioning

confidence: 99%

Forecasting and Anomaly Detection approaches using LSTM and LSTM Autoencoder techniques with the applications in supply chain management

Nguyen

Tran

Thomassey

et al. 2021

International Journal of Information Management

323

View full text Add to dashboard Cite

show abstract

“…With the continuous development of deep learning, many experts and scholars choose to use deep learning models such as RNN (Recurrent Neural Networks) and LSTM (Long Short-Term Memory) for anomaly detection [14,15]. For example, Hundman et al [16] used LSTM for anomaly detection in spatial telemetry time series data.…”

Section: Kpi Time Series Anomaly Detection Methodsmentioning

confidence: 99%

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

Shang

Zhang

et al. 2021

Applied Sciences

View full text Add to dashboard Cite

KPIs (Key Performance Indicators) in distributed systems may involve a variety of anomalies, which will lead to system failure and huge losses. Detecting KPI anomalies in the system is very important. This paper presents a time series anomaly detection method based on correlation analysis and HMM. Correlation analysis is used to obtain the correlation between abnormal KPIs in the system, thereby reducing the false alarm rate of anomaly detection. The HMM (Hidden Markov Model) is used for anomaly detection by finding the close relationship between abnormal KPIs. In our correlation analysis of abnormal KPIs, firstly, the time series prediction model (1D-CNN-TCN) is proposed. The residual sequence is obtained by calculating the residual between the predicted value and the actual value. The residual sequence can highlight the abnormal segment in each data point and improve the accuracy of anomaly screening. According to the obtained residual sequence, these abnormal KPIs are preliminarily screened out from the historical data. Next, KPI correlation analysis is performed, and the correlation score is obtained by adding a sliding window onto the obtained anomaly index residual sequence. The correlation analysis based on the residual sequence can eliminate the interference of the original data fluctuation itself. Then, a correlation matrix of abnormal KPIs is constructed using the obtained correlation scores. In anomaly detection, the constructed correlation matrix is processed to obtain the adaptive parameters of the HMM model, and the trained HMM is used to quickly discover the abnormal KPI that may cause a KPI anomaly. Experiments on public data sets show that the method obtains good results.

show abstract

“…For time series, LSTM auto-encoders capture timely dependencies [12]. The approach by Malhotra et al [12] is a semi-supervised autoencoder, that assumes access to purely normal training data, but auto-encoders can also be trained fully unsupervised on a mixture of normal and abnormal data [13]. With this approach, the assumption is that due to anomalies being underrepresented, if restricted sufficiently, the model will only fit the normal points and the reconstruction error will be larger for abnormal points or subsequences.…”

Section: Related Workmentioning

confidence: 99%

Outlier Detection with Explanations on Music Streaming Data: A Case Study with Danmark Music Group Ltd.

Sejr

Christiansen²,

Dvinge³

et al. 2021

Applied Sciences

View full text Add to dashboard Cite

In the digital marketplaces, businesses can micro-monitor sales worldwide and in real-time. Due to the vast amounts of data, there is a pressing need for tools that automatically highlight changing trends and anomalous (outlier) behavior that is potentially interesting to users. In collaboration with Danmark Music Group Ltd. we developed an unsupervised system for this problem based on a predictive neural network. To make the method transparent to developers and users (musicians, music managers, etc.), the system delivers two levels of outlier explanations: the deviation from the model prediction, and the explanation of the model prediction. We demonstrate both types of outlier explanations to provide value to data scientists and developers during development, tuning, and evaluation. The quantitative and qualitative evaluation shows that the users find the identified trends and anomalies interesting and worth further investigation. Consequently, the system was integrated into the production system. We discuss the challenges in unsupervised parameter tuning and show that the system could be further improved with personalization and integration of additional information, unrelated to the raw outlier score.

show abstract

Unsupervised Anomaly Detection in Time Series Using LSTM-Based Autoencoders

Cited by 99 publications

References 0 publications

Forecasting and Anomaly Detection approaches using LSTM and LSTM Autoencoder techniques with the applications in supply chain management

Forecasting and Anomaly Detection approaches using LSTM and LSTM Autoencoder techniques with the applications in supply chain management

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

Outlier Detection with Explanations on Music Streaming Data: A Case Study with Danmark Music Group Ltd.

Contact Info

Product

Resources

About