Open AccessForecasting tourist visits using data decomposition technique and learning optimization of artificial neural network
Author(s) -
S.Fis. Isnaini Herawati,
Muhammad Latif
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1811/1/012092
Subject(s) - artificial neural network , conjugate gradient method , tourism , decomposition , feedforward neural network , computer science , artificial intelligence , hilbert–huang transform , mean squared error , machine learning , data mining , statistics , algorithm , mathematics , geography , ecology , archaeology , filter (signal processing) , computer vision , biology
Estimates of tourist visits is very important to determining policy and decision making. This study proposed a new method for forecasting tourist visits. A case study was conducted at a tourist spot in Sumenep, Indonesia. The model proposed is data decomposition and optimization of learning against tourist visits data. Data decomposed use the Ensemble Empirical Mode Decomposition (EEMD) method, then data learning use the Feedforward Neural Network (FNN) which was optimized using the Polak-Ribiere Conjugate Gradient (PCG). The two methods are integrated to produce accurate forecasts. Several patterns of learning data were carried out in this experiment. The results of this method show good permformance results as measured used RMSE and MSE.