Deep learning for anomaly detection in multivariate time series: Approaches, applications, and challenges. Li, G. & Jung, J. J. Information Fusion, 91:93–102, March, 2023. ![Deep learning for anomaly detection in multivariate time series: Approaches, applications, and challenges [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper doi abstract bibtex Anomaly detection has recently been applied to various areas, and several techniques based on deep learning have been proposed for the analysis of multivariate time series. In this study, we classify the anomalies into three types, namely abnormal time points, time intervals, and time series, and review the state-of-the-art deep learning techniques for the detection of each of these types. Long short-term memory and autoencoders are the most commonly used methods for detecting abnormal time points and time intervals. In addition, some studies have implemented dynamic graphs to examine relational features between the time series and detect abnormal time intervals. However, anomaly detection still faces some limitations and challenges, such as the explainability of anomalies. Many studies have focused only on anomaly detection methods but failed to consider the reasons for the anomalies. Therefore, increasing the explainability of anomalies is an important research topic in anomaly detection.
@article{li_deep_2023,
title = {Deep learning for anomaly detection in multivariate time series: {Approaches}, applications, and challenges},
volume = {91},
issn = {1566-2535},
shorttitle = {Deep learning for anomaly detection in multivariate time series},
url = {https://www.sciencedirect.com/science/article/pii/S1566253522001774},
doi = {10.1016/j.inffus.2022.10.008},
abstract = {Anomaly detection has recently been applied to various areas, and several techniques based on deep learning have been proposed for the analysis of multivariate time series. In this study, we classify the anomalies into three types, namely abnormal time points, time intervals, and time series, and review the state-of-the-art deep learning techniques for the detection of each of these types. Long short-term memory and autoencoders are the most commonly used methods for detecting abnormal time points and time intervals. In addition, some studies have implemented dynamic graphs to examine relational features between the time series and detect abnormal time intervals. However, anomaly detection still faces some limitations and challenges, such as the explainability of anomalies. Many studies have focused only on anomaly detection methods but failed to consider the reasons for the anomalies. Therefore, increasing the explainability of anomalies is an important research topic in anomaly detection.},
urldate = {2023-11-14},
journal = {Information Fusion},
author = {Li, Gen and Jung, Jason J.},
month = mar,
year = {2023},
keywords = {Anomaly detection, Multivariate, Deep Learning},
pages = {93--102},
file = {1-s2.0-S1566253522001774-main.pdf:C\:\\Users\\Guillaume\\Zotero\\storage\\VIWT8ZWK\\1-s2.0-S1566253522001774-main.pdf:application/pdf},
}
Downloads: 0
{"_id":"EA6hfsNQ7fvQ3N67L","bibbaseid":"li-jung-deeplearningforanomalydetectioninmultivariatetimeseriesapproachesapplicationsandchallenges-2023","author_short":["Li, G.","Jung, J. J."],"bibdata":{"bibtype":"article","type":"article","title":"Deep learning for anomaly detection in multivariate time series: Approaches, applications, and challenges","volume":"91","issn":"1566-2535","shorttitle":"Deep learning for anomaly detection in multivariate time series","url":"https://www.sciencedirect.com/science/article/pii/S1566253522001774","doi":"10.1016/j.inffus.2022.10.008","abstract":"Anomaly detection has recently been applied to various areas, and several techniques based on deep learning have been proposed for the analysis of multivariate time series. In this study, we classify the anomalies into three types, namely abnormal time points, time intervals, and time series, and review the state-of-the-art deep learning techniques for the detection of each of these types. Long short-term memory and autoencoders are the most commonly used methods for detecting abnormal time points and time intervals. In addition, some studies have implemented dynamic graphs to examine relational features between the time series and detect abnormal time intervals. However, anomaly detection still faces some limitations and challenges, such as the explainability of anomalies. Many studies have focused only on anomaly detection methods but failed to consider the reasons for the anomalies. Therefore, increasing the explainability of anomalies is an important research topic in anomaly detection.","urldate":"2023-11-14","journal":"Information Fusion","author":[{"propositions":[],"lastnames":["Li"],"firstnames":["Gen"],"suffixes":[]},{"propositions":[],"lastnames":["Jung"],"firstnames":["Jason","J."],"suffixes":[]}],"month":"March","year":"2023","keywords":"Anomaly detection, Multivariate, Deep Learning","pages":"93–102","file":"1-s2.0-S1566253522001774-main.pdf:C\\:\\\\Users\\\\Guillaume\\\\Zotero\\\\storage\\\\VIWT8ZWK\\\\1-s2.0-S1566253522001774-main.pdf:application/pdf","bibtex":"@article{li_deep_2023,\n\ttitle = {Deep learning for anomaly detection in multivariate time series: {Approaches}, applications, and challenges},\n\tvolume = {91},\n\tissn = {1566-2535},\n\tshorttitle = {Deep learning for anomaly detection in multivariate time series},\n\turl = {https://www.sciencedirect.com/science/article/pii/S1566253522001774},\n\tdoi = {10.1016/j.inffus.2022.10.008},\n\tabstract = {Anomaly detection has recently been applied to various areas, and several techniques based on deep learning have been proposed for the analysis of multivariate time series. In this study, we classify the anomalies into three types, namely abnormal time points, time intervals, and time series, and review the state-of-the-art deep learning techniques for the detection of each of these types. Long short-term memory and autoencoders are the most commonly used methods for detecting abnormal time points and time intervals. In addition, some studies have implemented dynamic graphs to examine relational features between the time series and detect abnormal time intervals. However, anomaly detection still faces some limitations and challenges, such as the explainability of anomalies. Many studies have focused only on anomaly detection methods but failed to consider the reasons for the anomalies. Therefore, increasing the explainability of anomalies is an important research topic in anomaly detection.},\n\turldate = {2023-11-14},\n\tjournal = {Information Fusion},\n\tauthor = {Li, Gen and Jung, Jason J.},\n\tmonth = mar,\n\tyear = {2023},\n\tkeywords = {Anomaly detection, Multivariate, Deep Learning},\n\tpages = {93--102},\n\tfile = {1-s2.0-S1566253522001774-main.pdf:C\\:\\\\Users\\\\Guillaume\\\\Zotero\\\\storage\\\\VIWT8ZWK\\\\1-s2.0-S1566253522001774-main.pdf:application/pdf},\n}\n\n","author_short":["Li, G.","Jung, J. J."],"key":"li_deep_2023","id":"li_deep_2023","bibbaseid":"li-jung-deeplearningforanomalydetectioninmultivariatetimeseriesapproachesapplicationsandchallenges-2023","role":"author","urls":{"Paper":"https://www.sciencedirect.com/science/article/pii/S1566253522001774"},"keyword":["Anomaly detection","Multivariate","Deep Learning"],"metadata":{"authorlinks":{}},"html":""},"bibtype":"article","biburl":"https://bibbase.org/network/files/taiAEmmZxo42FqfQN","dataSources":["qG4exu4XyJLWyPN7m"],"keywords":["anomaly detection","multivariate","deep learning"],"search_terms":["deep","learning","anomaly","detection","multivariate","time","series","approaches","applications","challenges","li","jung"],"title":"Deep learning for anomaly detection in multivariate time series: Approaches, applications, and challenges","year":2023}