Comparative evaluation shows Random Forest provides best forecasting accuracy for tourism in Bali, indicating significant implications for planning and development.
Key Points
Random Forest achieved the lowest Root Mean Squared Error of 41772.68, indicating superior forecasting precision.
Hyperparameter optimization via Grid Search was applied to enhance the performance of machine learning models.
Forecasting models assessed include Random Forest, XGBoost, LSTM, and a Hybrid RF-XGBoost approach.
Results highlight the importance of considering seasonality and external shocks in tourism demand forecasting.