Open AccessAnalysis of the effect number input and hidden layer variations on the addition kohonen algorithm to backpropagation method
Author(s) -
Purwa Hasan Putra,
Muhammad Zarlis,
Herman Mawengkang
Publication year - 2020
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/725/1/012096
Subject(s) - backpropagation , epoch (astronomy) , computer science , self organizing map , algorithm , layer (electronics) , pattern recognition (psychology) , process (computing) , value (mathematics) , artificial neural network , artificial intelligence , data mining , machine learning , materials science , stars , composite material , computer vision , operating system
The process of recognizing data patterns using the method of adding kohonen to back propagation is very influential on the amount of input data and the number of hidden layers. The results of the testing with the addition of information on the back propagation algorithm have better epoch results in testing using input data 8 has an epoch value that is better than testing using the number of input data 3,4,5,6,7. The test results using layer 5 hidden have epoch 15 value which is better than the hidden layer 3, 4,6,7,8.