Prediksi Jumlah Perjalanan Wisatawan Nusantara Berdasarkan Data Historis Menggunakan Artificial Neural Network

Serenita Gisela Silalahi; Irene Lestaria sinaga; Stefani Silalahi

doi:10.37817/ikraith-informatika.v10i2.7109

Serenita Gisela Silalahi universitas hkbp nommensen pematang siantar
Irene Lestaria sinaga Universitas HKBP Nommensen Pematangsiantar
Stefani Silalahi Universitas HKBP Nommensen Pematangsiantar

DOI: https://doi.org/10.37817/ikraith-informatika.v10i2.7109

Keywords: Keywords: Artificial Neural Network, Backpropagation, Domestic Tourist, Historical Data, Prediction, Time Series Forecasting., Artificial Neural Network, Backpropagation, Domestic Tourist, Historical Data, Prediction, Time Series Forecasting.

Abstract

Domestic tourist travel is an important indicator of the development of Indonesia's domestic tourism sector. Predicting the number of domestic tourist trips is essential to support data-driven planning and decision-making. This study aims to develop a prediction model for domestic tourist travel based on historical data using an Artificial Neural Network (ANN). Secondary data obtained from Statistics Indonesia (BPS) were processed through preprocessing, Min-Max normalization, sliding window pattern formation, and ANN training using the Backpropagation algorithm. Model performance was evaluated using MAE, MSE, RMSE, MAPE, and R². The results obtained MAE of 0.0582, MSE of 0.0066, RMSE of 0.0813, MAPE of 9.44%, and R² of 0.3665, indicating that the proposed ANN model is capable of providing reasonably accurate predictions of domestic tourist travel based on historical data.