<![CDATA[The need for rice is an important factor in Indonesia where people make rice as a staple food source. Pematangsiantar city has a rice mill and rice sale, one of which is CV Hariara Pematangsiantar. Every year the amount of rice in CV Hariara Pematangsiantar changes. Therefore a prediction is needed to determine the amount of rice sales that will come and this prediction will be useful for the company to increase rice sales in the future. The data to be predicted is data on the amount of rice sales in CV Hariara Pematangsiantar in 2014-2017. The algorithm used to make predictions is a backpropagation neural network. There are five architectural models used in this prediction namely, 2-25-1 has an accuracy rate of 60%, 2-32-1=40%, 2-47-1=80%, 2-50-1=80%, and 2-52-1=60%. The best architecture of the five models is 2-47-1 with an accuracy of 80% and MSE of 7,46434101. So this architectural model is good enough to predict the amount of rice sales in CV Hariara]]>
Copyrights © 2021
