<![CDATA[Rainfall is a meteorological parameter that influences various sectors, such as agriculture, water resource management, and disaster mitigation; however, the process of classifying it still faces challenges, particularly due to imbalanced data across categories. This study aims to evaluate the performance of the XGBoost algorithm in classifying daily rainfall in the Special Region of Yogyakarta using NASA POWER data from 2000 to 2025, with input variables including air temperature, relative humidity, wind speed, and surface pressure. The evaluation was conducted using accuracy, precision, recall, and F1-score metrics to provide a more comprehensive overview of the model’s performance. The results indicate that the model achieved an accuracy of 0.82 and performed well in identifying light rain, and began to identify moderate rain, although not yet optimally; however, its performance remains limited for higher-intensity rain classes. This suggests that imbalanced data distribution remains a primary factor affecting model performance, making data quality and balance critical considerations in the development of rainfall classification models.]]>
Copyrights © 2026
