Comparative analysis using neural networks programming on Java for of signal recognition
DOI:
https://doi.org/10.34185/1562-9945-1-138-2022-18Keywords:
composite materials, neural networks, multilayer perceptron with back-propagation training, radial-basic neural network, defect, function of activityAbstract
The results of the study of a multilayer persertron and a radial-basic neural network for signal recognition are presented. Neural networks are implemented in Java in the environment NetBeans. The optimal number of neurons in the hidden layer is selected for building an effec-tive architecture of the neural network. Experiments were performed to analyze MSE values, Euclidean distance and accuracy.
References
Liu X. A review of artificial neural networks in the constitutive modeling of composite materials / Xin Liu, Su Tian, Fei Tao, Wenbin Yu // Composites Part B: Engineering , Elsevier, Nov. 2021, Vol. 224
Matveeva N. A. Surface defect detection with Neural Networks / N.A. Matveeva, A.A. Gurtovoy // System Technologies. – 2020.- Vol.126, No.1.- P. 96-102.
Fábio M. Soares, Alan M.F. Souza, Neural Network Programming with Java, - Bir-mingham, 2016. -244 p.
Haykin S. Neural Networks. A Comprehensive Foundation. Second edition. – New Jersey: Prentice Hall, 2008. -1103 p.
Herbert Schildt Java.The Complete Reference Ninth edition, 2014, 1372 p.
C. M. Bishop. Neural Networks for Pattern Recognition. Oxford University Press, 1995
James Levenick . Simply Java: An introduction to Java Programming. Charles River Media; 1st ed., September 8, 2005
