Abstract
Currency exchange is the exchange rate for current or future payments between two currencies of each country. In Indonesia, there are frequent fluctuations in the exchange rate of USD against IDR which causes instability in economic growth. This has resulted in reduced interest from foreign investors in investing in Indonesia, and has resulted in degeneration of development because the position of foreign investors is very important for economic growth. Therefore, predictions are needed to anticipate exchange rate fluctuations using the Long Short - Term Memory (LSTM) method. Some of the steps taken are collecting data, preprocessing, splitting data, build the LSTM model architecture, training the model, and testing. From the test results, the best results were obtained for the LSTM and LSTM + attention models, namely by using the parameters of 60 timestep, 32 neurons, 150 epoch, 32 batch size, and a learning rate of 0.001. The results obtained from the LSTM model are the total training time of 108.76 seconds, the loss value is 0.000162, and the RMSE result is 1.3328. The results obtained from the LSTM + attention model are the total training time of 116.05 seconds, the loss value is 0.000157, and the RMSE result is 0.6335. So it can be concluded that LSTM with attention can improve training accuracy.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
More From: Computatio : Journal of Computer Science and Information Systems
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.