PREDIKSI JUMLAH WISATAWAN MANCANEGARA KE INDONESIA MENGGUNAKAN ALGORITMA LINEAR REGRESSION DAN RANDOM FOREST REGRESSION

Abstract

Tourism is one of Indonesia’s leading sectors, contributing significantly to the national economy. Forecasting the number of international tourist arrivals is a strategic necessity to support policy planning and the sustainable development of the tourism industry. This study aims to compare the performance of two regression algorithms, Linear Regression and Random Forest Regression, in forecasting international tourist arrivals to Indonesia. The dataset covers the period 2020–2025, obtained from the Central Bureau of Statistics (BPS) with variables that underwent preprocessing such as normalization and handling of missing values. The methodology includes an 80:20 train-test split, tabular regression, and parameter tuning for Random Forest Regression to enhance model performance. Linear Regression was selected as a baseline model, while Random Forest Regression was chosen for its capability to model nonlinear patterns. Model evaluation was conducted using Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and R² Score. The results show that Linear Regression produced an RMSE of 59,967.668, MAE of 14,837.645, and R² Score of 0.007, indicating low accuracy. In contrast, Random Forest Regression achieved substantially better results with an RMSE of 9,696.530, MAE of 1,193.143, and R² Score of 0.974. These findings confirm that Random Forest Regression provides higher accuracy than Linear Regression, particularly in addressing seasonal patterns and uncertainties caused by global factors. In conclusion, Random Forest Regression can be considered a more reliable forecasting method for predicting international tourist arrivals. The forecasting results can serve as a basis for decision-making in destination capacity planning, foreign exchange revenue estimation, and risk mitigation against global disruptions (pandemics, geopolitical issues, crises), thereby supporting adaptive and sustainable strategies for national tourism development.