Peningkatan Nilai Akurasi Prediksi Algortima Backpropogation (Kasus: Jumlah Pengunjung Tamu pada Hotel berbintang di Sumatera Utara)
| Main Authors: | Syahfitri, Devi, Windarto, Agus Perdana, Fauzan, M, Solikhun, S |
|---|---|
| Format: | Article info application/pdf eJournal |
| Bahasa: | eng |
| Terbitan: |
Journal of Information System Research (JOSH)
, 2020
|
| Online Access: |
https://ejurnal.seminar-id.com/index.php/josh/article/view/476 https://ejurnal.seminar-id.com/index.php/josh/article/view/476/321 |
Daftar Isi:
- The research objective is to analyze and test whether the number of foreign tourist arrivals in Indonesia can be predicted using artificial intelligence techniques. The research data were obtained from the Indonesian Central Bureau of Statistics for the tourism category. The data used in the study is data on the number of tourist visits from 2003 - 2018. The artificial intelligence technique used in this study is the artificial neural network technique using the backpropagation method. The study will explore backpropagation parameters such as learning rate and network architecture where the results of the study are expected to help provide information for decision-makers to attract more foreign tourists to Indonesia because this has an impact on improving the economy in Indonesia. The research results show that from 4 architectural models tested (7-2-1, 7-5-1, 7-10-1 and 7-2-10-1) with a learning rate of 0.1; 0.01; 0.001; 0.2; 0.02; 0.002; 0.3; 0.03; and 0.003 where the 7-10-1 model with learning rate = 0.02 is obtained which is the best prediction model with the Root Mean Squared Error is 0.094 and the relative error is 4.49% +/- 10.21%. The accuracy of the truth obtained is 96%.