Clustering of Data Streams With Dynamic Gaussian Mixture Models: An IoT Application in Industrial Processes

Díaz-Rozo, Javier; Bielza, Concha; Larrañaga, Pedro

doi:10.1109/jiot.2018.2840129

Cited by 67 publications

(25 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…As a result, it has been demonstrated that unsupervised ML algorithms embedded in CPS are the key enablers for working towards highly accurate diagnosis and prognosis tools. In another study proposed by the same authors [18], they presented a new unsupervised learning algorithm based on GMM called Gaussian-based dynamic This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/ This article has been accepted for publication in a future issue of this journal, but has not been fully edited.…”

Section: Related Workmentioning

confidence: 99%

Multi-Machine Gaussian Topic Modeling for Predictive Maintenance

2021

View full text Add to dashboard Cite

In this paper, we propose a coherent framework for multi-machine analysis, using a group clustering model, which can be utilized for predictive maintenance (PdM). The framework benefits from the repetitive structure posed by multiple machines and enables for assessment of health condition, degradation modeling and comparison of machines. It is based on a hierarchical probabilistic model, denoted Gaussian topic model (GTM), where cluster patterns are shared over machines and therefore it allows one to directly obtain proportions of patterns over the machines. This is then used as a basis for cross comparison between machines where identified similarities and differences can lead to important insights about their degradation behavior. The framework is based on aggregation of data over multiple streams by a predefined set of features extracted over a time window. Moreover, the framework contains a clustering schema which takes uncertainty of cluster assignments into account and where one can specify a desirable degree of reliability of the assignments. By using a multi-machine simulation example, we highlight how the framework can be utilized in order to obtain cluster patterns and inherent variations of such patterns over machines. Furthermore, a comparative study with the commonly used Gaussian mixture model (GMM) demonstrates that GTM is able to identify inherent patterns in the data while the GMM fails. Such result is a consequence of the group level being modeled by the GTM while being absent in the GMM. Hence, the GTM are trained with a view on the data that is not available to the GMM with the consequence that the GMM can miss important, possibly even key, cluster patterns. Therefore, we argue that more advanced cluster models, like the GTM, can be key for interpreting and understanding degradation behavior across machines and ultimately for obtaining more efficient and reliable PdM systems.INDEX TERMS exploratory data analysis, cluster analysis, Gaussian topic modeling, hierarchical modeling, multi-machine analysis, multiple data streams, predictive maintenance

show abstract

Section: Related Workmentioning

confidence: 99%

Multi-Machine Gaussian Topic Modeling for Predictive Maintenance

2021

View full text Add to dashboard Cite

show abstract

“…Kullanıcıya ait click verisi analizi [6], saldırı tespit sistemleri [7][8][9], sosyal medya [10][11][12], finansal uygulamalar [13], bilimsel araştırmalar [14], sağlık araştırmaları [15][16][17], mobil uygulamalar [18], nesnelerin interneti (IoT) [19] ve sensor ağ [20,21] gibi pek çok alanda kullanılmaktadır. Nesnelerin interneti konusunun yaygınlaştığı günümüzde uygulama alanlarının daha da artacağını söylemek mümkündür.…”

Section: Akan Veri Kümeleme Yaklaşımlarının Uygulama Alanlarıunclassified

Akan Veri Kümeleme Teknikleri Üzerine Bir Derleme

Şenol

Karacan

2018

European Journal of Science and Technology

View full text Add to dashboard Cite

Öz Günümüz teknolojisinin gelişmesine paralel olarak bilgisayar ortamına aktarılmış olan veri miktarı inanılmaz boyutlara ulaşmış ve gün geçtikçe de artmaktadır. Bu nedenle veriyi işleme yöntemleri de değişmektedir. Klasik kümeleme yaklaşımlarında veri statiktir. Oysa günümüz teknolojisinde, verinin çok hızlı olduğu dünyada artık veriyi akarken kümeleyecek, kullanıcıya istediği zaman sonuç verebilecek uygulamalara ihtiyaç vardır. Bu anlamda ihtiyacı karşılayan akan veri kümeleme yaklaşımlarına olan talep gün geçtikçe artmaktadır. Çünkü akan veri kümeleme yaklaşımları bir defa okumalı, hızlı ve kendisini yeni gelen veriye uyarlama özelliğine sahiptir. Yani veri bir yandan akarken bir yandan kullanıcıya sonuç üretilebilmektedir. Bu çalışmada akan veri kümeleme alanında yapılan çalışmalar derlenmekte ve bu alana ilgi duyan araştırmacılara ışık tutulmaktır.

show abstract

“…In addition, the algorithm operates with very small amounts of data and greatly reduces calculation power to determine whether to modify the model. e algorithm can be evaluated on synthetic data and data stream from a test bed where various operating conditions are detected automatically, with good results in terms of classification precision, sensitivity, and characteristics [2]. Based on trajectory data, regular behaviour of private cars are extracted [3][4][5].…”

Section: Introductionmentioning

confidence: 99%

Big Data and Its V’s with IoT to Develop Sustainability

Chu

Nazir

Wang

et al. 2021

Scientific Programming

View full text Add to dashboard Cite

Internet of Things (IoT) has been considered as one of the emerging network and information technologies that can comprehend automatic monitoring, identification, and management through a network of smart IoT devices. The effective use of IoT in different areas has improved efficiency and reduced errors. The rapid growth of smart devices such as actuators, sensors, and wearable devices has made the IoT enable for smart and sustainable developments in the area. Physical objects are interlinked with these smart devices for the progression to analyse, process, and manage the surroundings data. Such data can then be further utilised for smarter decisions and postanalysis for different purposes. However, with the limited IoT resources, the management of data is difficult due to the restrictions of transmission power place and energy consumption, and the processing can put pressure on these smart devices. The network of IoT is connected with big data through Internet for manipulating and storing huge bulk of data on cloud storage. The secure framework based on big data through IoT is the awful need of modern society which can be energy efficient in a sustainable environment. Due to the intrinsic characteristics of sensors nodes in the IoT, like data redundancy, constrained energy, computing capabilities, and limited communication range, the issues of data loss are becoming among the main issues which mostly depend on the completeness of data. Various approaches are in practice for the recovery problem of data, such as spatiotemporal correlation and interpolation. These are used for data correlation and characteristics of sensory data. Extracting correlation data became difficult specifically as the coupling degree between diverse perceptual attributes is low. The current study has presented a comprehensive overview on big data and its V’s with Internet of Things to describe the research into the area with in-depth review of existing literature.

show abstract

Clustering of Data Streams With Dynamic Gaussian Mixture Models: An IoT Application in Industrial Processes

Cited by 67 publications

References 33 publications

Multi-Machine Gaussian Topic Modeling for Predictive Maintenance

Multi-Machine Gaussian Topic Modeling for Predictive Maintenance

Akan Veri Kümeleme Teknikleri Üzerine Bir Derleme

Big Data and Its V’s with IoT to Develop Sustainability

Contact Info

Product

Resources

About