Perbandingan Sistem Prediksi Cuaca Harian Menggunakan Algoritma SVR dan XGBOOST

Abstract

Mountain climbing activities are highly influenced by dynamic weather conditions; therefore, a system capable of accurately predicting daily weather is essential. This study focuses on comparing the performance of two machine learning algorithms — Support Vector Regression (SVR) and Extreme Gradient Boosting (XGBoost) — in developing a daily weather prediction system. Weather data were obtained from the Meteorology, Climatology, and Geophysics Agency (BMKG) with parameters including minimum temperature, maximum temperature, average temperature, humidity, rainfall, solar radiation, maximum wind speed, and wind direction. The research stages include data collection, preprocessing, model training using SVR and XGBoost, and performance evaluation using Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Accuracy, and R² metrics. The results show that the SVR algorithm performs better than XGBoost in predicting daily weather data, particularly for temperature and humidity variables, with more stable accuracy across various observation stations.