Adaptive learning of an evolving cascade neo-fuzzy system in data stream mining tasks

Bodyanskiy, Yevgeniy; Tyshchenko, Oleksii K.; Kopaliani, Daria S.

doi:10.1007/s12530-016-9149-5

Cited by 31 publications

(12 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For the simulation part, four systems were chosen that had been previously developed by us since they could demonstrate quite impressive results from our previous research. For this reason, we took a CasNN network (Bodyanskiy et al, 2015b), a CasENFN system (Bodyanskiy et al, 2016), an EvoENFN system (Hu et al, 2017), and the recent hybrid cascade neuro-fuzzy network (HybCas) (Bodyanskiy and Tyshchenko, 2020). To add a few third-party evolving systems to our comparison, two more systems were chosen (eNFN (Silva et al, 2013) and eMG (Caminhas et al, 2011)).…”

Section: Simulation Resultsmentioning

confidence: 99%

“…A modified procedure by Kaczmarz-Widrow-Hoff (Kaczmarz, 1937) was applied for training this system. In the work of Bodyanskiy et al (2016), a similar cascade hybrid network was built on the basis of extended NFNs (ENFNs) (Hu et al, 2017) for the first time. At the same time, a new procedure for optimization of a neurons pool was also presented in that paper.…”

Section: Structure Of the Hybrid Cascade Neuro-fuzzy Network With Poomentioning

confidence: 99%

See 1 more Smart Citation

A Hybrid Cascade Neuro–Fuzzy Network with Pools of Extended Neo–Fuzzy Neurons and its Deep Learning

Bodyanskiy

Tyshchenko

2019

International Journal of Applied Mathematics and Computer Science

Self Cite

View full text Add to dashboard Cite

This research contribution instantiates a framework of a hybrid cascade neural network based on the application of a specific sort of neo-fuzzy elements and a new peculiar adaptive training rule. The main trait of the offered system is its competence to continue intensifying its cascades until the required accuracy is gained. A distinctive rapid training procedure is also covered for this case that offers the possibility to operate with non-stationary data streams in an attempt to provide online training of multiple parametric variables. A new training criterion is examined for handling non-stationary objects. Additionally, there is always an occasion to set up (increase) the inference order and the number of membership relations inside the extended neo-fuzzy neuron.

show abstract

Section: Simulation Resultsmentioning

confidence: 99%

Section: Structure Of the Hybrid Cascade Neuro-fuzzy Network With Poomentioning

confidence: 99%

A Hybrid Cascade Neuro–Fuzzy Network with Pools of Extended Neo–Fuzzy Neurons and its Deep Learning

Bodyanskiy

Tyshchenko

2019

International Journal of Applied Mathematics and Computer Science

Self Cite

View full text Add to dashboard Cite

show abstract

“…Dynamic clustering has been tackled by several authors. Among the main methods we can mention connectionist approaches like Self-Adaptive Feed-Forward Neural Networks (SAFN) [19] and the evolving cascade connectionist system of [20], fuzzy logic approaches among which Evolving Clustering [17,21], the algorithm pClass to extract fuzzy rules [22] and also [23,24], neo-fuzzy approaches [25], Growing Gaussian Mixture Models (2G2M) [26], and random forests [27]. Deep learning architectures have also been proposed [28].…”

Section: Related Workmentioning

confidence: 99%

: Dynamic clustering for tracking evolving environments

Roa

Travé-Massuyès

Grisales

2019

Pattern Recognition

View full text Add to dashboard Cite

Evolving environments challenge researchers with non stationary data flows where the concepts-or states-being tracked can change over time. This requires tracking algorithms suited to represent concept evolution and in some cases, e.g. real industrial environments, also suited to represent time dependent features. This paper proposes a unified approach to track evolving environments that uses a two-stages distance-based and density-based clustering algorithm. In this approach data samples are fed as input to the distance based clustering stage in an incremental, online fashion, and they are then clustered to form µ-clusters. The density-based algorithm analyses the micro-clusters to provide the final clusters: thanks to a forgetting process, clusters may emerge, drift, merge, split or disappear, hence following the evolution of the environment. This algorithm has proved to be able to detect high overlapping clusters even in multi-density distributions, making no assumption about cluster convexity. It shows fast response to data streams and good outlier rejection properties.

show abstract

“…Virtual management is considered as the basis for the application of basic provisions in the development of infor-mation communication technologies (ICT), specifically for transportation processes [5,6]. Study [5] employs fuzzy logic for this purpose.…”

Section: Literature Review and Problem Statementmentioning

confidence: 99%

“…Study [5] employs fuzzy logic for this purpose. Paper [6] addresses solution to the problem of application of neuromathematics for transport applications. General theoretical and applied problems of information support for this development are given in results of [7,8].…”

Section: Literature Review and Problem Statementmentioning

confidence: 99%

Development of automotive computer systems based on the virtualization of transportation processes management

Aleksiyev

Klets

et al. 2017

EEJET

View full text Add to dashboard Cite

show abstract

Adaptive learning of an evolving cascade neo-fuzzy system in data stream mining tasks

Cited by 31 publications

References 24 publications

A Hybrid Cascade Neuro–Fuzzy Network with Pools of Extended Neo–Fuzzy Neurons and its Deep Learning

A Hybrid Cascade Neuro–Fuzzy Network with Pools of Extended Neo–Fuzzy Neurons and its Deep Learning

: Dynamic clustering for tracking evolving environments

Development of automotive computer systems based on the virtualization of transportation processes management

Contact Info

Product

Resources

About