LSTM Learning with Bayesian and Gaussian Processing for Anomaly Detection in Industrial IoT

Di Wu; Zhongkai Jiang; Xiaofeng Xie; Xuetao Wei; Weiren Yu; Renfa Li

doi:10.1109/TII.2019.2952917

LSTM Learning with Bayesian and Gaussian Processing for Anomaly Detection in Industrial IoT

Di Wu, Zhongkai Jiang, Xiaofeng Xie, Xuetao Wei, Weiren Yu, Renfa Li

College of Engineering and Physical Sciences

Research output: Contribution to journal › Article › peer-review

Abstract

The data generated by millions of sensors in Industrial Internet of Things (IIoT) is extremely dynamic, heterogeneous, and large scale. It poses great challenges on the real-time analysis and decision making for anomaly detection in IIoT. In this paper, we propose a LSTM-Gauss-NBayes method, which is a synergy of the long short-term memory neural network (LSTM-NN) and the Gaussian Bayes model for outlier detection in IIoT. In a nutshell, the LSTM-NN builds model on normal time series. It detects outliers by utilising the predictive error for the Gaussian Naive Bayes model. Our method exploits advantages of both LSTM and Gaussian Naive Bayes models, which not only has strong prediction capability of LSTM for future time point data, but also achieves an excellent classification performance of Gaussian Naive Bayes model through the predictive error. Empirical studies demonstrate our solution outperforms the best-known competitors, which is a preferable choice for detecting anomalies.

Original language	English
Pages (from-to)	5244-5253
Number of pages	10
Journal	IEEE Transactions on Industrial Informatics
Volume	16
Issue number	8
Early online date	11 Nov 2019
DOIs	https://doi.org/10.1109/TII.2019.2952917
Publication status	Published - Aug 2020

Bibliographical note

© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

Funding: HuXiang Youth Talent Program; National Natural Science Foundation of China.

Access to Document

10.1109/TII.2019.2952917

LSTM Learning with Bayesian and Gaussian Processing
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Accepted author manuscript, 1.53 MB

Cite this

@article{9221e122bf3c4999a805840476c0d0ed,

title = "LSTM Learning with Bayesian and Gaussian Processing for Anomaly Detection in Industrial IoT",

abstract = "The data generated by millions of sensors in Industrial Internet of Things (IIoT) is extremely dynamic, heterogeneous, and large scale. It poses great challenges on the real-time analysis and decision making for anomaly detection in IIoT. In this paper, we propose a LSTM-Gauss-NBayes method, which is a synergy of the long short-term memory neural network (LSTM-NN) and the Gaussian Bayes model for outlier detection in IIoT. In a nutshell, the LSTM-NN builds model on normal time series. It detects outliers by utilising the predictive error for the Gaussian Naive Bayes model. Our method exploits advantages of both LSTM and Gaussian Naive Bayes models, which not only has strong prediction capability of LSTM for future time point data, but also achieves an excellent classification performance of Gaussian Naive Bayes model through the predictive error. Empirical studies demonstrate our solution outperforms the best-known competitors, which is a preferable choice for detecting anomalies.",

author = "Di Wu and Zhongkai Jiang and Xiaofeng Xie and Xuetao Wei and Weiren Yu and Renfa Li",

note = "{\textcopyright} 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. Funding: HuXiang Youth Talent Program; National Natural Science Foundation of China. ",

year = "2020",

month = aug,

doi = "10.1109/TII.2019.2952917",

language = "English",

volume = "16",

pages = "5244--5253",

journal = "IEEE Transactions on Industrial Informatics",

issn = "1551-3203",

publisher = "IEEE",

number = "8",

}

TY - JOUR

T1 - LSTM Learning with Bayesian and Gaussian Processing for Anomaly Detection in Industrial IoT

AU - Wu, Di

AU - Jiang, Zhongkai

AU - Xie, Xiaofeng

AU - Wei, Xuetao

AU - Yu, Weiren

AU - Li, Renfa

N1 - © 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. Funding: HuXiang Youth Talent Program; National Natural Science Foundation of China.

PY - 2020/8

Y1 - 2020/8

N2 - The data generated by millions of sensors in Industrial Internet of Things (IIoT) is extremely dynamic, heterogeneous, and large scale. It poses great challenges on the real-time analysis and decision making for anomaly detection in IIoT. In this paper, we propose a LSTM-Gauss-NBayes method, which is a synergy of the long short-term memory neural network (LSTM-NN) and the Gaussian Bayes model for outlier detection in IIoT. In a nutshell, the LSTM-NN builds model on normal time series. It detects outliers by utilising the predictive error for the Gaussian Naive Bayes model. Our method exploits advantages of both LSTM and Gaussian Naive Bayes models, which not only has strong prediction capability of LSTM for future time point data, but also achieves an excellent classification performance of Gaussian Naive Bayes model through the predictive error. Empirical studies demonstrate our solution outperforms the best-known competitors, which is a preferable choice for detecting anomalies.

AB - The data generated by millions of sensors in Industrial Internet of Things (IIoT) is extremely dynamic, heterogeneous, and large scale. It poses great challenges on the real-time analysis and decision making for anomaly detection in IIoT. In this paper, we propose a LSTM-Gauss-NBayes method, which is a synergy of the long short-term memory neural network (LSTM-NN) and the Gaussian Bayes model for outlier detection in IIoT. In a nutshell, the LSTM-NN builds model on normal time series. It detects outliers by utilising the predictive error for the Gaussian Naive Bayes model. Our method exploits advantages of both LSTM and Gaussian Naive Bayes models, which not only has strong prediction capability of LSTM for future time point data, but also achieves an excellent classification performance of Gaussian Naive Bayes model through the predictive error. Empirical studies demonstrate our solution outperforms the best-known competitors, which is a preferable choice for detecting anomalies.

UR - https://ieeexplore.ieee.org/document/8896029/

U2 - 10.1109/TII.2019.2952917

DO - 10.1109/TII.2019.2952917

M3 - Article

SN - 1551-3203

VL - 16

SP - 5244

EP - 5253

JO - IEEE Transactions on Industrial Informatics

JF - IEEE Transactions on Industrial Informatics

IS - 8

ER -

LSTM Learning with Bayesian and Gaussian Processing for Anomaly Detection in Industrial IoT

Abstract

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this