PREDIKSI PRODUKTIVITAS JAGUNG DI INDONESIA SEBAGAI UPAYA ANTISIPASI IMPOR MENGGUNAKAN JARINGAN SARAF TIRUAN BACKPROPAGATION

Abstract

Corn is a staple food that is still widely consumed by the population of Indonesia. Based on data from the Indonesian Statistics Agency, corn productivity in Indonesia from 2005 to 2015 calculated an unstable curve. Therefore this research was conducted to predict and see the large growth of maize in Indonesia for the following years so that the government has a reference to continuously strive to increase corn productivity in Indonesia in order to remain stable in order to meet the needs of Indonesian people to minimize corn imports. This study uses data on corn productivity in Indonesia in 2005-2015 sourced from the Indonesian Central Bureau of Statistics. The prediction algorithm used is the Backpropagation Neural Network. This algorithm is able to predict data well, especially data that is maintained for a certain period of time. To facilitate data analysis, the author uses the Matlab 2011b application. In this study, a training and testing process will be carried out using 5 network architecture models, namely 5-25-1, 5-43-1, 5-76-1, 5-78-1 and 7-128-1. Of the 5 architectural models, the best is 5-25-1 with the percentage of 88% and the MSE value of 0.00992433.