Machine Learning for Flood Prediction in Indonesia: Providing Online Access for Disaster Management Control

AbstractAs one of the most vulnerable countries to floods, there should be an increased necessity for accurate and reliable flood forecasting in Indonesia. Therefore, a new prediction model using machine learning algorithm is proposed provide daily Data crawling was conducted obtain rainfall, streamflow, land cover, data from 2008 2021. The built Random Forest (RF) classification predict future floods by inputting three days rainfall rate, forest ratio, stream flow. accuracy, specificity, precision, recall, F1-score on test dataset RF are approximately 94.93%, 68.24%, 94.34%, 99.97%, 97.08%, respectively. Moreover, AUC (Area Under Curve) ROC (Receiver Operating Characteristics) curve results 71%. objective this research providing that predicts events accurately Indonesian regions 3 months prior day flood. As trial, we used month June 2022 predicted accurately. result then published website as warning system form mitigation.Keywords : algorithm, random forest, online access website, mitigation, F-score