PERAMALAN NILAI TUKAR KURS RUPIAH TERHADAP DOLLAR AMERIKA MENGGUNAKAN METODE HEBB NEURAL NETWORK

Abstract

Artifical neural networks have been widely used to help solve a wide range of issues, one of the problems is the use of artificial Hebb Neural Network to classify forecasting the value of the Rupiah exchange rate to Dollar America. This application uses a maximum of 10.000 literation, learning rate of 0.6 and a target error of 0.0001. The test result of 56 data is the result accuracy of 96.63% or 8.33% error rate. The error may occur because the neural network if there is a similar training data would be difficult to recognize patterns. Keywords: Hebb Neural Network, Java Netbeans.