Powered by OpenAIRE graph
Found an issue? Give us feedback
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/ YÖK Açık Bilim - CoH...arrow_drop_down
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
versions View all 2 versions
addClaim

Multivariate time series clustering using variable order markov models and its applications on cyber-physical systems

Authors: Sürmeli, Barış Gün;

Multivariate time series clustering using variable order markov models and its applications on cyber-physical systems

Abstract

Siber-Fiziksel Sistemler'den elde edilen Çok-değişkenli Zaman Serileri (CZS) verisi, sistemin karakteristik özellikleri hakkında değerli bilgiler içermektedir. Bir Makine Öğrenmesi yöntemi olan, Çok-değişkenli Zaman Serileri (CZS) Kümelemesi, sistemin değişik çalışma aralıklarında gösterdiği davranışların arasındaki benzerlikleri açığa çıkarmak için kullanılabilir. Sistem hakkındaki bu bilgiler, hata tespiti, sistem bakımı ve kök neden analizi gibi görevlerin gerçekleştirilmesi için ön bilgi sağlayabilir. Bu tezde, her bir CZS'yi, istatistiksel bir yöntem olan Değişken Dereceli Markov Zincirleri (DDMZ) ile modellenmiş, ve elde edilen bu modelleri karşılaştırarak aralarındaki uzaklıkları/benzerlikleri hesaplamak için kullanılmak üzere yeni bir metrik sunulmuştur. Elde edilen bu ikili uzaklıklar baz alınarak DDMZ'ler kümelendirilmiş ve bu şekilde CZS Kümelemesi görevi sonuçlandırılmıştır. Biri Gizli Markov Modelleri, diğeri ise Temel Bileşenler Analizi kullanarak CZS'leri modelleyen iki yöntem karşılaştırma amacıyla açıklanmıştır. Sunulan yöntemin üstünlüğü, biri siber-fiziksel laboratuvar göstericisinden elde edilmiş, diğeri ise endüstriyel bulaşık makinesi üretim fabrikasından elde edilmiş iki veri seti üzerinde yapılan deneylerle doğrulanmıştır. Ayrıca, yeni bir DDMZ öğrenme yöntemi sunulmuş ve üç CZS Kümeleme yöntemi için hesaplama karmaşıklığı tartışılmıştır.

Multivariate Time Series (MTS) data obtained from Cyber-Physical Systems carry resourceful information about the internal characteristics of the system. As one of the exploratory Machine Learning methods, Multivariate Time Series Clustering can enable one to discover the similarities and differences of the manifested behavior in different working periods/cycles of a system. This information can then be used as a prior knowledge for tasks such as anomaly detection, system maintenance or root-cause analysis. In this thesis, we make use of the statistical method, Variable Order Markov Models (VOMMs) to model each individual MTS and present a new metric to calculate the distances between those VOMMs. The VOMMs are then clustered with respect to these pairwise distances to complete the MTS Clustering task. Two other MTS Clustering methods which use Hidden Markov Models and Principal Component Analysis to model the MTSs are also explained. The superiority of the proposed method is confirmed with the experiments on two data sets; one obtained from a cyber-physical lab demonstrator and one from an industrial dishwasher production plant. A new VOMM construction method as well as the computational complexity of the three MTS Clustering methods are also discussed.

77

Country
Turkey
Related Organizations
Keywords

Computer engineering, Bilgisayar mühendisliği, Computer Engineering and Computer Science and Control, Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve Kontrol

  • BIP!
    Impact byBIP!
    selected citations
    These citations are derived from selected sources.
    This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    0
    popularity
    This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
    Average
    influence
    This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    Average
    impulse
    This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
    Average
Powered by OpenAIRE graph
Found an issue? Give us feedback
selected citations
These citations are derived from selected sources.
This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Citations provided by BIP!
popularity
This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
BIP!Popularity provided by BIP!
influence
This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Influence provided by BIP!
impulse
This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
BIP!Impulse provided by BIP!
0
Average
Average
Average
Green
Funded by
Related to Research communities
RISIS