<![CDATA[Flooding is one of the most frequent hydrometeorological disasters in Indonesia, highlighting the need for an early warning system capable of delivering timely and accurate information to support disaster mitigation efforts. This study aims to design and develop an Android-based flood early warning application by implementing the Random Forest algorithm as a classification method based on weather data. The research data were obtained from the Indonesian Agency for Meteorology, Climatology, and Geophysics (BMKG) over a six-month period, including air temperature, humidity, rainfall, and wind speed. The dataset was processed through preprocessing, data labeling, and dataset splitting using an 80:20 ratio for training and testing data, respectively. The evaluation results showed that the proposed model achieved an accuracy of 96.97%, a precision of 97.98%, a recall of 96.97%, and an F1-score of 97.06%, indicating excellent classification performance. The trained model was then integrated into an Android application capable of providing weather information, flood potential predictions, and early warning notifications based on the user's location. The results demonstrate that the developed application functions according to the system requirements and has the potential to enhance community preparedness by providing flood potential information that is timely, easily accessible, and informative.]]>
Copyrights © 2026
