Prediksi Jumlah Nilai Impor Sumatera Utara Menurut Negara Asal Menggunakan Algoritma Backpropagation

In fulfilling domestic needs, the government must conduct international economic
activities such as imports. Import is the process of transporting goods or commodities from one
country to another legally. Particularly in the province of North Sumatra there is always an
increase in the number of imports annually starting in 2014-2016 at the Central Bureau of
Statistics (BPS) of North Sumatra. In this research, the writer will predict the value of import
value for 5 years ahead by using backpropagation algorithm. Backpropagation is one method of
Artificial Neural Network (Artificial Neural Network), which is quite reliable in solving
problems. One is the prediction of the amount of import value in North Sumatra. This study uses
5 architectural models: 4-12-1, 4-15-1,4-18-1, 4-19-1, 4-20-1, of the five models, are the best
accuracy obtained from the architectural model 4-19 -1 with 100% accuracy value, epoch 2807
iteration, and MSE is 0.00099930653.