Simulasi Algoritma Jaringan Syaraf Tiruan Order-Satu

Abstrak

Abstract
Artificial neural network is a computational algorithm that mimics the workings of nerve cells.
All the incoming signal is multiplied by a weight that is on each input, by neuronal cells, all
signals have been multiplied by weights are summed and then added again with bias. The
sum is input to a function (activation function) produces the output of neurons (here used a
linear activation function). During the learning process, the weights and biases are always
updated using learning algorithms. if there is an error in the output. For the identification
process, the weights are directly weighing input is what is called as a search parameter, the
price of w1, w2, w3 and w4. In the on-line identification, neurons or networks of neurons will
always be 'learned' every input and output data.
Keywords: neuro, activation function, learning constants.