Using parallel random search to train fuzzy neural networks

Oliinyk, Аndrii; Skrupsky, Stepan; Subbotin, Sergey

doi:10.3103/s0146411614060078

Cited by 13 publications

(5 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Than we may determine the fuzzy inference system of the Mamdani-Zadeh classifier type, which in the form of a neuro-fuzzy network can be further trained by means of optimization methods [54][55][56] to adjust the parameters of membership functions to fuzzy terms and weights of rules that provide acceptable values of the clustering quality functional.…”

Section: Methodsmentioning

confidence: 99%

“…Let evaluate the performance quality of a NFN (the quality of data clustering) based on a given functional F, which can be determined based on a wide class of metrics [1,2,7,8,[12][13][14][15]. Using the methods of evolutionary optimization [54,55], we can select such values of the network term parameters that will improve the value of the optimized functional F. The final model will be a neuro-fuzzy clusterizer optimized by the number of features used and the functional F.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Data Clustering Based on Inductive Learning of Neuro-Fuzzy Network With Distance Hashing

Субботин

2022

RIC

View full text Add to dashboard Cite

Context. Cluster analysis is widely used to analyze data of various nature and dimensions. However, the known methods of cluster analysis are characterized by low speed and are demanding on computer memory resources due to the need to calculate pairwise distances between instances in a multidimensional feature space. In addition, the results of known methods of cluster analysis are difficult for human perception and analysis with a large number of features. Objective. The purpose of the work is to increase the speed of cluster analysis, the interpretability of the resulting partition into clusters, as well as to reduce the requirements of cluster analysis to computer memory. Method. A method for cluster analysis of multidimensional data is proposed, which for each instance calculates its hash based on the distance to the conditional center of coordinates, uses a one-dimensional coordinate along the hash axis to determine the distances between instances, considers the resulting hash as a pseudo-output feature, breaking it into intervals, which matches the labels pseudo-classes – clusters, having received a rough crisp partition of the feature space and sample instances, automatically generates a partition of input features into fuzzy terms, determines the rules for referring instances to clusters and, as a result, forms a fuzzy inference system of the Mamdani-Zadeh classifier type, which is further trained in the form of a neuro-fuzzy network to ensure acceptable values of the clustering quality functional. This makes it possible to reduce the number of terms and features used, to evaluate their contribution to making decisions about assigning instances to clusters, to increase the speed of data cluster analysis, and to increase the interpretability of the resulting data splitting into clusters. Results. The mathematical support for solving the problem of cluster data analysis in conditions of large data dimensions has been developed. The experiments confirmed the operability of the developed mathematical support have been carried out. Conclusions. . The developed method and its software implementation can be recommended for use in practice in the problems of analyzing data of various nature and dimensions.

show abstract

Section: Methodsmentioning

confidence: 99%

mentioning

confidence: 99%

Data Clustering Based on Inductive Learning of Neuro-Fuzzy Network With Distance Hashing

Субботин

2022

RIC

View full text Add to dashboard Cite

show abstract

“…On the rules of selection on every iteration there are only those individuals that is most adjusted-exactly they are used by the operator of crossing, for the receipt of descendants. Sometimes after crossing there are mutations-casual changes in a chromosome, that allow at the decision of tasks optimizations to go out from a local minimum [1][2][3][9][10][11][12].…”

Section: Classification Analysis Of the Functioning Of The Investigated Methodsmentioning

confidence: 99%

“…For n points exists (n − 1) possible ways. For example, for 10 points, the number of paths is more than 3 million, and for 20 points it is several quintillions [10,11].…”

Section: Development Of An Evolutionary Methods For Solving the Traveling Salesman Problemmentioning

confidence: 99%

Implementation of Evolutionary Methods of Solving the Travelling Salesman Problem in a Robotic Warehouse

Oliinyk

Fedorchenko

Stepanenko

et al. 2020

Data-Centric Business and Applications

View full text Add to dashboard Cite

An evolutionary method for solving the traveling salesman problem in the field of pharmacy business by optimizing the work of the drug delivery device is proposed in this paper. Modifications of three methods of initialization of the initial population of the genetic algorithm are developed. The software implementation is proposed to solve the problem of a sales-man in the pharmacy business by optimizing the process of drug delivery, using modified evolutionary methods. Unlike existing methods, the modified version of the evolution method allows to choose the original method of population initialization when solving the problem of traveling salesman, which, in turn, allows to generate more adapted chromosomes (chromosomes with better values of fitness functionality) at the stage of initialization and thereby improve the results algorithm. It is also possible to graphically monitor the process of solving the seller's problem, get the result in text and graphic forms. The principles of objectoriented programming, namely the use of classes, the principle of data encapsulation and inheritance, were used when writing the program. UML diagrams of classes, sequences, activities, states and cooperation were used to visualize the structure and functional relationships of the modules of the developed software.

show abstract

“…Shevchuk, V. P. Metrologiya intellektual'nyh izmeritel'nyh sistem [Text]: monografiya / V. P. Shevchuk, V. I. Kaplya [5][6][7], Big Data [8], Artificial Intelligence [9], Internet of Things [10]. For instance, the synergy of SDN and Neural Networks technologies can be expedient in the context of parallel computing systems' resources allocation [11,12].…”

Section: Introductionmentioning

confidence: 99%

Development of stratified approach to software defined networks simulation

Shkarupylo

Skrupsky

Oliinyk

et al. 2017

EEJET

View full text Add to dashboard Cite

Запропоновано стратифікований підхід до імітаційного моделювання програмно-конфі-гурованих мереж. Запропоновано імітаційні моделі мережі, активних і пасивних компо-нентів -контролера, комутатора, хоста та комунікаційних каналів. Придатність підхо-ду до цільового використання підтверджено шляхом співставлення одержаних результа-тів імітаційного моделювання із результата-ми емуляції мережі у середовищі Mininet Ключові слова: програмно-конфігурована мережа, імітаційне моделювання, дискрет-но-подійна специфікація системи, великі дані Предложен стратифицированный подход к имитационному моделированию программно-конфигурируемых сетей. Предложены ими-тационные модели сети, активных и пассив-ных компонентов -контроллера, коммута-тора, хоста и коммуникационных каналов. Пригодность подхода к целевому использо-ванию подтверждена путем сопоставления полученных результатов имитационного мо-делирования с результатами эмулирования сети в среде Mininet Ключевые слова: программно-конфигури-руемая сеть, имитационное моделирование, дискретно-событийная спецификация систе-мы, большие данные UDC 004.75 : 004.94

show abstract

Using parallel random search to train fuzzy neural networks

Cited by 13 publications

References 5 publications

Data Clustering Based on Inductive Learning of Neuro-Fuzzy Network With Distance Hashing

Data Clustering Based on Inductive Learning of Neuro-Fuzzy Network With Distance Hashing

Implementation of Evolutionary Methods of Solving the Travelling Salesman Problem in a Robotic Warehouse

Development of stratified approach to software defined networks simulation

Contact Info

Product

Resources

About