Development of an algorithm to train artificial neural networks for intelligent decision support systems
Authors:
- Oleg SovaOSOleg Sova
- Oleksandr TurinskyiOTOleksandr Turinskyi
- Andrii ShyshatskyiASAndrii Shyshatskyi
- Volodymyr DudnykVDVolodymyr Dudnyk
- Ruslan ZhyvotovskyiRZRuslan Zhyvotovskyi
- Yevgen ProkopenkoYPYevgen Prokopenko
- Taras HurskyiTHTaras Hurskyi
- Valerii HordiichukVHValerii Hordiichuk
- Anton NikitenkoANAnton Nikitenko
- Artem RemezARArtem Remez
Open Access
- 3 February 2020
- journal article
- research article
- Published by Private Company Technology Center in Eastern-European Journal of Enterprise Technologies
- Vol. 1 (9 (103)) , 46-55
Abstract
The algorithm to train artificial neural networks for intelligent decision support systems has been constructed. A distinctive feature of the proposed algorithm is that it conducts training not only for synaptic weights of an artificial neural network, but also for the type and parameters of membership function. In case of inability to ensure the assigned quality of functioning of artificial neural networks due to training of parameters of artificial neural network, the architecture of artificial neural networks is trained. The choice of the architecture, type and parameters of membership function occurs taking into consideration the computation resources of the facility and taking into consideration the type and the amount of information entering the input of an artificial neural network. In addition, when using the proposed algorithm, there is no accumulation of an error of artificial neural networks training as a result of processing the information entering the input of artificial neural networks.Development of the proposed algorithm was predetermined by the need to train artificial neural networks for intelligent decision support systems in order to process more information given the unambiguity of decisions being made. The research results revealed that the specified training algorithm provides on average 16–23 % higher the efficiency of training artificial neural networks training that is on average by 16–23 % higher and does not accumulate errors in the course of training. The specified algorithm will make it possible to conduct training of artificial neural networks; to determine effective measures to enhance the efficiency of functioning of artificial neural networks. The developed algorithm will also enable the improvement of the efficiency of functioning of artificial neural networks due to training the parameters and the architecture of artificial neural networks. The proposed algorithm reduces the use of computational resources of decision support systems. The application of the developed algorithm makes it possible to work out the measures aimed at improving the effectiveness of training artificial neural networks and to increase the efficiency of information processingKeywords
This publication has 24 references indexed in Scilit:
- An Artificially Intelligent Technique to Generate Synthetic Geomechanical Well Logs for the Bakken FormationEnergies, 2018
- Airline customer lifetime value estimation using data analytics supported by social network informationJournal of Air Transport Management, 2018
- Anomaly detection in wireless sensor network of the “smart home” system2017
- Prediction of Air Pollutants Concentration Based on an Extreme Learning Machine: The Case of Hong KongInternational Journal of Environmental Research and Public Health, 2017
- Flexible Neo-fuzzy Neuron and Neuro-fuzzy Network for Monitoring Time Series PropertiesInformation Technology and Management Science, 2013
- Evolving Connectionist Systems2003
- Self-Organizing Maps1995
- Fuzzy basis functions, universal approximation, and orthogonal least-squares learningIEEE Transactions on Neural Networks, 1992
- Identification and control of dynamical systems using neural networksIEEE Transactions on Neural Networks, 1990
- Structure identification of fuzzy modelFuzzy Sets and Systems, 1988