<![CDATA[Rainfall is a crucial meteorological factor that profoundly impacts various aspects of life, particularly in tropical regions such as Manado City, where fluctuations in precipitation have significant consequences across sectors ranging from agriculture and water resource management to the potential for hydrometeorological disasters like floods and landslides. Accurate rainfall prediction is therefore essential, yet the inherent complexity of tropical atmospheric systems often poses considerable challenges for traditional forecasting methods. This research explores the potential of machine learning to enhance the precision of daily rainfall predictions in Manado City. We implement two distinct machine learning algorithms, namely Support Vector Machine (SVM) and Gated Recurrent Unit (GRU), to forecast rainfall quantities based on historical meteorological parameters obtained from the Indonesian Agency for Meteorology, Climatology, and Geophysics (BMKG). The dataset utilized encompasses vital variables such as air temperature, relative humidity, wind speed, and atmospheric pressure. The primary objective of this study is to develop, evaluate, and compare the performance of these two models in predicting rainfall, with a specific focus on their capacity to capture complex patterns and dependencies within time-series data. The models will be trained using historical data and rigorously assessed based on standard performance metrics, including Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and the coefficient of determination (R2). It is anticipated that the findings from this research will offer valuable insights into the effectiveness of machine learning algorithms for local rainfall forecasting, thereby contributing to the development of more reliable early warning systems and improved climate adaptation strategies for the community of Manado]]>
