Paper published in a book (Scientific congresses and symposiums)
An Anomaly Detection Approach based on the combination of LSTM Autoencoder and Isolation Forest for multivariate time series data
Tran, Phuong Hanh; Heuchenne, Cédric; Thomassey, Sébatien
2020In 14th International FLINS Conference on Robotics and Artificial Intelligence, Germany 2020
Peer reviewed
 

Files


Full Text
autoLSTM_iforest.pdf
Author preprint (332.65 kB)
Request a copy

All documents in ORBi are protected by a user license.

Send to



Details



Keywords :
Autoencoder, Isolation Forest; Long Short Term Memory, Time series data; Anomaly Detection
Abstract :
[en] It is true that anomaly detection is an important issue that has had a long history in the research community due to its various applications. Literature has recorded various Artificial Intelligence (AI) techniques that have been applied to detect anomalies without having a priori knowledge about them. Anomaly detection approaches for multivariate time series data have still too many unrealistic assumptions to apply to the industry. Our paper, therefore, proposed a new efficiency approach of anomaly detection for multivariate time series data. We specifically developed a new hybrid approach based on LSTM Autoencoder and Isolation Forest (iForest). This approach enables the advantages in extracting good features of the LSTM Autoencoder and the good performance in anomaly detection problems of the iForest. The results show that our approach leads to the improvement of performance significantly in comparison with the One-Class Support Vector Machine (OCSVM) method. Our approach is implemented on simulated data in the fashion industry (FI).
Disciplines :
Computer science
Author, co-author :
Tran, Phuong Hanh  ;  Université de Liège - ULiège > HEC Recherche
Heuchenne, Cédric ;  Université de Liège - ULiège > HEC Liège : UER > UER Opérations: Statistique appl. à la gest. et à l'économie
Thomassey, Sébatien
Language :
English
Title :
An Anomaly Detection Approach based on the combination of LSTM Autoencoder and Isolation Forest for multivariate time series data
Publication date :
2020
Event name :
14th International FLINS Conference on Robotics and Artificial Intelligence
Event date :
from 18-08-2020 to 21-08-2020
Audience :
International
Main work title :
14th International FLINS Conference on Robotics and Artificial Intelligence, Germany 2020
Edition :
FLINS 2020 proceedings
Peer reviewed :
Peer reviewed
Available on ORBi :
since 15 June 2020

Statistics


Number of views
198 (14 by ULiège)
Number of downloads
9 (2 by ULiège)

Bibliography


Similar publications



Contact ORBi