Jaringan Saraf Tiruan Backpropagation Untuk Memprediksi Jumlah Ekspor Buah-Buahan Menurut Negara Tujuan

Nurhidayana Nurhidayana (STIKOM Tunas Bangsa, Pematangsiantar)
Irfan Sudahri Damanik (STIKOM Tunas Bangsa, Pematangsiantar)
Rafiqa Dewi (AMIK Tunas Bangsa, Pematangsiantar)

JOURNAL OF INFORMATION SYSTEM RESEARCH (JOSH)
Vol 2 No 2 (2021): Januari 2021

This research aims to predict the number of exports of fruits by country of destination in the coming year based on data in the previous year. The application of a prediction is very important in conducting research. The method used in conducting this prediction is the artificial neural network (JST) with its Backpropagation algorithm. The architecture used in this study is 4-2-1, 4-4-1, 4-6-1, 4-8-1, 4-4-2-1, and 4-3-6-1. Test results obtained prediction of export of fruits according to destination country with 91% accuracy rate with 4-4-1 architecture