A Novel Hybrid Method for KPI Anomaly Detection Based on VAE and SVDD

Key performance indicator (KPI) anomaly detection is the underlying core technology in Artificial Intelligence for IT operations (AIOps). It has an important impact on subsequent anomaly location and root cause analysis. Variational auto-encoder (VAE) is a symmetry network structure composed of enco...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Yun Zhao, Xiuguo Zhang, Zijing Shang, Zhiying Cao
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	key performance indicator (KPI) anomaly detection variational auto-encoder (VAE) support vector data description (SVDD) Mathematics QA1-939
Acceso en línea:	https://doaj.org/article/1d74e5d79e2f4370b034f2cec1600548
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:1d74e5d79e2f4370b034f2cec1600548
record_format	dspace
spelling	oai:doaj.org-article:1d74e5d79e2f4370b034f2cec16005482021-11-25T19:06:46ZA Novel Hybrid Method for KPI Anomaly Detection Based on VAE and SVDD10.3390/sym131121042073-8994https://doaj.org/article/1d74e5d79e2f4370b034f2cec16005482021-11-01T00:00:00Zhttps://www.mdpi.com/2073-8994/13/11/2104https://doaj.org/toc/2073-8994Key performance indicator (KPI) anomaly detection is the underlying core technology in Artificial Intelligence for IT operations (AIOps). It has an important impact on subsequent anomaly location and root cause analysis. Variational auto-encoder (VAE) is a symmetry network structure composed of encoder and decoder, which has attracted extensive attention because of its ability to capture complex KPI data features and better detection results. However, VAE is not well applied to the modeling of KPI time series data and it is often necessary to set the threshold to obtain more accurate results. In response to these problems, this paper proposes a novel hybrid method for KPI anomaly detection based on VAE and support vector data description (SVDD). This method consists of two modules: a VAE reconstructor and SVDD anomaly detector. In the VAE reconstruction module, firstly, bi-directional long short-term memory (BiLSTM) is used to replace the traditional feedforward neural network in VAE to capture the time correlation of sequences; then, batch normalization is used at the output of the encoder to prevent the disappearance of <i>KL</i> (Kullback–Leibler) divergence, which prevents ignoring latent variables to reconstruct data directly. Finally, exponentially weighted moving average (EWMA) is used to smooth the reconstruction error, which reduces false positives and false negatives during the detection process. In the SVDD anomaly detection module, smoothed reconstruction errors are introduced into the SVDD for training to determine the threshold of adaptively anomaly detection. Experimental results on the public dataset show that this method has a better detection effect than baseline methods.Yun ZhaoXiuguo ZhangZijing ShangZhiying CaoMDPI AGarticlekey performance indicator (KPI)anomaly detectionvariational auto-encoder (VAE)support vector data description (SVDD)MathematicsQA1-939ENSymmetry, Vol 13, Iss 2104, p 2104 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	key performance indicator (KPI) anomaly detection variational auto-encoder (VAE) support vector data description (SVDD) Mathematics QA1-939
spellingShingle	key performance indicator (KPI) anomaly detection variational auto-encoder (VAE) support vector data description (SVDD) Mathematics QA1-939 Yun Zhao Xiuguo Zhang Zijing Shang Zhiying Cao A Novel Hybrid Method for KPI Anomaly Detection Based on VAE and SVDD
description	Key performance indicator (KPI) anomaly detection is the underlying core technology in Artificial Intelligence for IT operations (AIOps). It has an important impact on subsequent anomaly location and root cause analysis. Variational auto-encoder (VAE) is a symmetry network structure composed of encoder and decoder, which has attracted extensive attention because of its ability to capture complex KPI data features and better detection results. However, VAE is not well applied to the modeling of KPI time series data and it is often necessary to set the threshold to obtain more accurate results. In response to these problems, this paper proposes a novel hybrid method for KPI anomaly detection based on VAE and support vector data description (SVDD). This method consists of two modules: a VAE reconstructor and SVDD anomaly detector. In the VAE reconstruction module, firstly, bi-directional long short-term memory (BiLSTM) is used to replace the traditional feedforward neural network in VAE to capture the time correlation of sequences; then, batch normalization is used at the output of the encoder to prevent the disappearance of <i>KL</i> (Kullback–Leibler) divergence, which prevents ignoring latent variables to reconstruct data directly. Finally, exponentially weighted moving average (EWMA) is used to smooth the reconstruction error, which reduces false positives and false negatives during the detection process. In the SVDD anomaly detection module, smoothed reconstruction errors are introduced into the SVDD for training to determine the threshold of adaptively anomaly detection. Experimental results on the public dataset show that this method has a better detection effect than baseline methods.
format	article
author	Yun Zhao Xiuguo Zhang Zijing Shang Zhiying Cao
author_facet	Yun Zhao Xiuguo Zhang Zijing Shang Zhiying Cao
author_sort	Yun Zhao
title	A Novel Hybrid Method for KPI Anomaly Detection Based on VAE and SVDD
title_short	A Novel Hybrid Method for KPI Anomaly Detection Based on VAE and SVDD
title_full	A Novel Hybrid Method for KPI Anomaly Detection Based on VAE and SVDD
title_fullStr	A Novel Hybrid Method for KPI Anomaly Detection Based on VAE and SVDD
title_full_unstemmed	A Novel Hybrid Method for KPI Anomaly Detection Based on VAE and SVDD
title_sort	novel hybrid method for kpi anomaly detection based on vae and svdd
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/1d74e5d79e2f4370b034f2cec1600548
work_keys_str_mv	AT yunzhao anovelhybridmethodforkpianomalydetectionbasedonvaeandsvdd AT xiuguozhang anovelhybridmethodforkpianomalydetectionbasedonvaeandsvdd AT zijingshang anovelhybridmethodforkpianomalydetectionbasedonvaeandsvdd AT zhiyingcao anovelhybridmethodforkpianomalydetectionbasedonvaeandsvdd AT yunzhao novelhybridmethodforkpianomalydetectionbasedonvaeandsvdd AT xiuguozhang novelhybridmethodforkpianomalydetectionbasedonvaeandsvdd AT zijingshang novelhybridmethodforkpianomalydetectionbasedonvaeandsvdd AT zhiyingcao novelhybridmethodforkpianomalydetectionbasedonvaeandsvdd
_version_	1718410264333778944

A Novel Hybrid Method for KPI Anomaly Detection Based on VAE and SVDD

Ejemplares similares