Determination of Accuracy at Backpropagation Method in Prediction Crude Oil Prices

Jhon Veri, Surmayanti Surmayanti, Guslendra Guslendra

© 2021 Jhon Veri, published by UIKTEN. This work is licensed under the Creative Commons Attribution-NonCommercial 4.0 International. (CC BY-NC 4.0).

Citation Information: SAR Journal. Volume 4, Issue 4, Pages 181-184, ISSN 2619-9955, https://doi.org/10.18421/SAR44-05, December 2021.

Received: 17 November 2021.
Revised:   16 December 2021.
Accepted: 20 December 2021.
Published: 27 December 2021.

Abstract:

We analyzed the performance of the artificial neural network with the backpropagation method in predicting crude oil prices in this paper, including the case of crude oil price predictions. The training results obtained that the MSE value was 0.00099762 with 135 Epoch, in the network testing the MSE value was 0.093336. Meanwhile, the predicted value is determined by the target value with a contribution of 99% with a significant effect. Thus the accuracy level is determined by the target value and the predicted value. The accuracy of the system is obtained for 83,6%.

Keywords – Backpropagation, Accuracy, Prediction.

                   

                                                                        Full text PDF