Design of Reinforced Interval Type-2 Fuzzy C-Means-Based Fuzzy Classifier

Kim, Eun-Hu; Oh, Sung‐Kwun; Pedrycz, Witold

doi:10.1109/tfuzz.2017.2785244

Cited by 53 publications

(12 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In this section, the performance of the proposed E-ELM for the optimisation of the antecedent and consequent parts of the IT2-RBFNN having a fixed mean m si and a variable standard deviation [σ 1 i , σ 2 i ] is compared to other techniques such as an IT2-RBFNN trained with an Adaptive Gradient Descent (AGD) approach [5,23,26], Support Vector Machines (SVMs), Back propagation networks (BPNs), RBFNN and an Interval Type-2 Fuzzy Neural Network with Support Vector Regression. We use two complex data sets from the UCI repository, i.e.…”

Section: Resultsmentioning

confidence: 99%

“…Fuzzy Logic Systems (FLSs) have been widely used to solve a large number of real world problems [1][2][3][4][5]. In particular, in the areas of function approximation and classification problems, FLSs of Interval Type-2 (IT2) have demonstrated to be more efficient to handle with uncertainties, such as nosiy and sparse data as well as their ability to operate under disturbances that usually T1 FLSs can not [5,6].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Evolutionary Extreme Learning Machine for the Interval Type-2 Radial Basis Function Neural Network: A Fuzzy Modelling Approach

Rubio-Solis

Martinez-Hernandez

Panoutsos

2018

2018 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE)

View full text Add to dashboard Cite

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Evolutionary Extreme Learning Machine for the Interval Type-2 Radial Basis Function Neural Network: A Fuzzy Modelling Approach

Rubio-Solis

Martinez-Hernandez

Panoutsos

2018

2018 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE)

View full text Add to dashboard Cite

show abstract

“…Content may change prior to final publication. largest, but the similarity between different categories the smallest [42]. When FCM is applied to the clustering of WSN nodes, the network nodes can be divided into several tight clusters to minimize the transmission distance when the members send data to the cluster head, so as to reduce the communication energy consumption within the cluster.…”

Section: ) Fcm Clustering Algorithmmentioning

confidence: 99%

An Energy Efficient Routing Protocol Based on Improved Artificial Bee Colony Algorithm for Wireless Sensor Networks

et al. 2020

View full text Add to dashboard Cite

Clustering sensor nodes is an effective method in designing routing algorithms for Wireless Sensor Networks (WSNs), which improves network lifetime and energy efficiency. In clustered WSNs, cluster heads are the key nodes, they need to perform more tasks, so they consume more energy. Therefore, it is an important problem to select the optimal cluster heads. In this paper, we propose a clustering algorithm that selects cluster heads using an improved artificial bee colony (ABC) algorithm. Based on the standard ABC algorithm, an efficient improved ABC algorithm is proposed, and then the network cluster head energy, cluster head density, cluster head location and other similar factors are introduced into the improved ABC algorithm theory to solve the clustering problem in WSNs. In the network initialization period, all nodes have the same energy level, the improved ABC algorithm is used to optimize fuzzy C-means clustering to find the optimal clustering method. We also propose an energy-efficient routing algorithm based on an improved ant colony optimization for routing between the cluster heads and the base station. In order to improve energy efficiency and further improve network throughput, in the stable transmission phase, we introduce a polling control mechanism based on busy/idle nodes into intra-cluster communication. The performance of the proposed protocol is evaluated in several different scenarios. The simulation results show that the proposed protocol has a better performance compared to a number of recent similar protocols. INDEX TERMS WSN, clustering, energy efficiency, network lifetime, high throughput, polling, routing algorithm, artificial bee colony

show abstract

“…2.1 Estimação Paramétrica do Antecedente O particionamento dos dados experimentais, de acordo com a metodologia adotada, implicará na definição das regiões de operação e, necessariamente, do número de regras do FKF. Dentre os diversos algoritmos existentes para esta finalidade, o FCM apresenta importante aplicabilidade devido a sua eficiência e simplicidade de implementação (Zhang et al, 2019;Kim et al, 2018). O algoritmo de agrupamento FCMé formulado como segue (Babuska, 1998).…”

Section: Estimação Paramétrica Do Filtro De Kalman Fuzzyunclassified

Projeto de Filtro de Kalman Fuzzy Baseado em Modelo via Decomposição Espectral de Dados Experimentais

Serra

Gomes

2019

Anais Do 14º Simpósio Brasileiro De Automação Inteligente

View full text Add to dashboard Cite

In this paper, a methodology for design of fuzzy Kalman filter (FKF) via spectral decomposition of the experimental data, is proposed. The adopted methodology consists in parametric estimation of local state space linear submodels of a fuzzy model of the dynamic system, by means of a fuzzy algorithm based on least squares, as well as in estimation of FKF gains from the fuzzy model, using the parallel and distributed compensation (PDC) method. The partitioning of experimental data, for definition of the rule base as well as the nonlinear FKF characteristic, is performed by fuzzy C-Means (FCM) clustering algorithm. Considering the PDC method, the Kalman gains in the consequent of each FKF rule are updated as a function of the unobservable components resulting from the spectral decomposition of noisy experimental data. In particular, it allows to FKF efficiency in the face of considerable variations in the noise level over the experimental data (signal/noise ratio). Computational results illustrate the good performance of the proposed methodology when compared to relevant approaches from the literature. Resumo: Neste artigoé proposta uma metodologia para projeto de filtro de Kalman fuzzy (FKF) via decomposição espectral dos dados experimentais. A metodologia adotada consiste na estimação paramétrica dos submodelos lineares locais de um modelo fuzzy no espaço de estados, referente ao sistema dinâmico, por meio de um algoritmo fuzzy baseado em mínimos quadrados. A estimação dos ganhos do FKF baseado no modelo obtidoé realizada pelo método de compensação paralela e distribuída (CPD). O particionamento dos dados experimentaisé realizado pelo algoritmo fuzzy C-Means (FCM) para definição da base de regras, bem como da característica não linear do FKF. Considerando-se o método CPD, os ganhos de Kalman no consequente de cada regra do FKF são atualizados em função das componentes nãoobserváveis resultantes da decomposição espectral do sinal ruidoso. Este aspecto, em particular, permite a eficiência do FKF diante de variações consideráveis do nível de ruído sobre os dados experimentais (relação sinal/ruído). Resultados computacionais ilustram o bom desempenho da metodologia proposta quando comparada a abordagens relevantes da literatura.

show abstract

Design of Reinforced Interval Type-2 Fuzzy C-Means-Based Fuzzy Classifier

Cited by 53 publications

References 32 publications

Evolutionary Extreme Learning Machine for the Interval Type-2 Radial Basis Function Neural Network: A Fuzzy Modelling Approach

Evolutionary Extreme Learning Machine for the Interval Type-2 Radial Basis Function Neural Network: A Fuzzy Modelling Approach

An Energy Efficient Routing Protocol Based on Improved Artificial Bee Colony Algorithm for Wireless Sensor Networks

Projeto de Filtro de Kalman Fuzzy Baseado em Modelo via Decomposição Espectral de Dados Experimentais

Contact Info

Product

Resources

About