Pemanfaatan ANN untuk Prediksi Penjualan Online Industri Rumahan selama Pandemi Covid-19

Abstract

Research has been conducted on the use of ANN for predicting online sales of the home industry during the Covid-19 pandemic. The purpose of this study is to train and test neural networks with the backpropagation method to obtain accurate forecasting results. The data were obtained by providing a list of questions to the home industry. The data obtained is separated into 2 parts, the first part is used to train the artificial neural network, the second part is used to test the performance of the artificial neural network. From the training that has been carried out, the success of finding goals according to the predetermined error tolerance value is 0.02 at the 33 epoch, input layer is 5 neurons, the hidden layer is 3 neurons, error tolerance is 0.025, learning rate is equal to 0.1. From the test, we managed to find a goal according to the predetermined error tolerance value, which is 0.2 at the 100 epoch, the input layer is 5 neurons, the hidden layer is 4 neurons, error tolerance is 0.2, the learning rate is equal to 0.1. The conclusion is that the artificial neural network can perform training and testing well, so as to produce precise and accurate predictions, the results of these predictions can be taken into consideration for making a decision.