Implementasi Metode Backpropagation Neural Network Dalam Memprediksi Hasil Produksi Kedelai

Abstract

The necessary of soybean in Indonesia tends to increase annually. However, soybean production tends to decrease so that soybean needs does not fullfilled. One of the environmental factors that influence soybean production is climate such as temperature, humidity, sunlight, rainfall, and wind velocity. This study aims to predict soybean production results based on the influence of climate by using an Artificial Neural Network (ANN) method. The algorithm used is Backpropagation with climate and soybean production results in the previous period parameters as input in the prediction process. The results of this study get a training accuracy of 96.6% and testing accuracy of 96.5%.