Multivariate Time Series Anomaly Detection Using Working Memory Connections in Bi-Directional Long Short-Term Memory Autoencoder Network
“Normal” events are characterized as data patterns or behaviors that align with expected operational conditions, while “anomalies” are defined as deviations from these patterns, potentially signaling faults, errors, or unexpected system behaviors. The timely and accurate detection of anomalies plays...
| Main Authors: | , , , |
|---|---|
| 格式: | 文件 |
| 语言: | English |
| 出版: |
MDPI AG
2025-03-01
|
| 丛编: | Applied Sciences |
| 主题: | |
| 在线阅读: | https://www.mdpi.com/2076-3417/15/5/2861 |
| _version_ | 1826531793917968384 |
|---|---|
| author | Xianghua Ding Jingnan Wang Yiqi Liu Uk Jung |
| author_facet | Xianghua Ding Jingnan Wang Yiqi Liu Uk Jung |
| author_sort | Xianghua Ding |
| collection | DOAJ |
| description | “Normal” events are characterized as data patterns or behaviors that align with expected operational conditions, while “anomalies” are defined as deviations from these patterns, potentially signaling faults, errors, or unexpected system behaviors. The timely and accurate detection of anomalies plays a critical role in domains such as industrial manufacturing, financial transactions, and other related domains. In the context of Industry 4.0, the proliferation of sensors has resulted in a massive influx of time series data, making the anomaly detection of such multivariate time series data a popular research area. Long Short-Term Memory (LSTM) has been extensively recognized as an effective framework for modeling and processing time series data. Previous studies have combined Bi-directional Long Short-Term Memory (Bi-LSTM) architecture with Autoencoder (AE) for multivariate time series anomaly detection. However, due to the inherent limitations of LSTM, Bi-LSTM-AE still cannot overcome these drawbacks. Our study replaces the LSTM units within the Bi-LSTM-AE architecture of existing research with Working Memory Connections for LSTM units and demonstrates that this architecture performs better in the field of multivariate time series anomaly detection compared to using standard LSTM units. The model we proposed not only outperforms the baseline models but also demonstrates greater robustness across various scenarios. |
| first_indexed | 2025-03-14T01:40:59Z |
| format | Article |
| id | doaj.art-916aea900f7f42e28216d2102066136d |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2025-03-14T01:40:59Z |
| publishDate | 2025-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-916aea900f7f42e28216d2102066136d2025-03-12T13:54:08ZengMDPI AGApplied Sciences2076-34172025-03-01155286110.3390/app15052861Multivariate Time Series Anomaly Detection Using Working Memory Connections in Bi-Directional Long Short-Term Memory Autoencoder NetworkXianghua Ding0Jingnan Wang1Yiqi Liu2Uk Jung3Department of Business Administration, School of Business, Dongguk University-Seoul, Seoul 04620, Republic of KoreaDepartment of Business Administration, School of Business, Dongguk University-Seoul, Seoul 04620, Republic of KoreaDepartment of Business Administration, School of Business, Dongguk University-Seoul, Seoul 04620, Republic of KoreaDepartment of Business Administration, School of Business, Dongguk University-Seoul, Seoul 04620, Republic of Korea“Normal” events are characterized as data patterns or behaviors that align with expected operational conditions, while “anomalies” are defined as deviations from these patterns, potentially signaling faults, errors, or unexpected system behaviors. The timely and accurate detection of anomalies plays a critical role in domains such as industrial manufacturing, financial transactions, and other related domains. In the context of Industry 4.0, the proliferation of sensors has resulted in a massive influx of time series data, making the anomaly detection of such multivariate time series data a popular research area. Long Short-Term Memory (LSTM) has been extensively recognized as an effective framework for modeling and processing time series data. Previous studies have combined Bi-directional Long Short-Term Memory (Bi-LSTM) architecture with Autoencoder (AE) for multivariate time series anomaly detection. However, due to the inherent limitations of LSTM, Bi-LSTM-AE still cannot overcome these drawbacks. Our study replaces the LSTM units within the Bi-LSTM-AE architecture of existing research with Working Memory Connections for LSTM units and demonstrates that this architecture performs better in the field of multivariate time series anomaly detection compared to using standard LSTM units. The model we proposed not only outperforms the baseline models but also demonstrates greater robustness across various scenarios.https://www.mdpi.com/2076-3417/15/5/2861anomaly detectionlong short-term memorytime series dataautoencodermultivariate time series |
| spellingShingle | Xianghua Ding Jingnan Wang Yiqi Liu Uk Jung Multivariate Time Series Anomaly Detection Using Working Memory Connections in Bi-Directional Long Short-Term Memory Autoencoder Network Applied Sciences anomaly detection long short-term memory time series data autoencoder multivariate time series |
| title | Multivariate Time Series Anomaly Detection Using Working Memory Connections in Bi-Directional Long Short-Term Memory Autoencoder Network |
| title_full | Multivariate Time Series Anomaly Detection Using Working Memory Connections in Bi-Directional Long Short-Term Memory Autoencoder Network |
| title_fullStr | Multivariate Time Series Anomaly Detection Using Working Memory Connections in Bi-Directional Long Short-Term Memory Autoencoder Network |
| title_full_unstemmed | Multivariate Time Series Anomaly Detection Using Working Memory Connections in Bi-Directional Long Short-Term Memory Autoencoder Network |
| title_short | Multivariate Time Series Anomaly Detection Using Working Memory Connections in Bi-Directional Long Short-Term Memory Autoencoder Network |
| title_sort | multivariate time series anomaly detection using working memory connections in bi directional long short term memory autoencoder network |
| topic | anomaly detection long short-term memory time series data autoencoder multivariate time series |
| url | https://www.mdpi.com/2076-3417/15/5/2861 |
| work_keys_str_mv | AT xianghuading multivariatetimeseriesanomalydetectionusingworkingmemoryconnectionsinbidirectionallongshorttermmemoryautoencodernetwork AT jingnanwang multivariatetimeseriesanomalydetectionusingworkingmemoryconnectionsinbidirectionallongshorttermmemoryautoencodernetwork AT yiqiliu multivariatetimeseriesanomalydetectionusingworkingmemoryconnectionsinbidirectionallongshorttermmemoryautoencodernetwork AT ukjung multivariatetimeseriesanomalydetectionusingworkingmemoryconnectionsinbidirectionallongshorttermmemoryautoencodernetwork |