Open AccessAnalisis JST Dalam Memprediksi Jumlah Tamu Pada Hotel NonBintang
Author(s) -
Bil Klinton Sihotang,
Anjar Wanto
Publication year - 2018
Publication title -
techno.com
Language(s) - English
Resource type - Journals
eISSN - 2356-2579
pISSN - 1412-2693
DOI - 10.33633/tc.v17i4.1762
Subject(s) - tourism , hospitality , backpropagation , artificial neural network , government (linguistics) , computer science , operations research , business , engineering , geography , artificial intelligence , linguistics , philosophy , archaeology
Analysis on a prediction (forecasting) is very important to do in a study, So with this data analysis will be obtained a clear picture of the issues discussed. As well as in predicting the number of guests in non-star hotels. This research is expected to be useful for both Government and private parties as one of the study materials in the development of hotel business, as well as for academics as study material / research especially related to tourism and hospitality field. The data used in this study is data on the number of guests in non-star hotels by province from the Central Bureau of Statistics Indonesia from 2011 to 2016. This study uses the method of artificial neural network Backpropagation using 5 architectural models, those are 4-19-1, 4-50-1, 4-17-1, 4-16-1, 4-22-. From architecture, the best architecture is 12-19-1 with an accuracy of 88.2%, MSE 0.10206089 with error rate used 0.001 - 0.05. Thus, this model is good enough to predict the number of guests indonesia in non-star hotels