Flooding is a significant concern in nations with frequent precipitation because it can instantly affect multiple regions simultaneously. Due to the unpredictability of their occurrence caused by rapid water level rise, it is challenging to predict such natural disasters accurately. During flooding, prompt rescue efforts are crucial for the affected population. Due to flooded highways and residences, rescue teams may have difficulty locating victims. This hinders the potentially perilous and time-consuming rescue operation. To address this problem, we propose a web-based system that integrates natural language processing (NLP) with global positioning system (GPS) functionality. The SOSFloodFinder system provides automatic classification priorities for text messages sent by flood victims, as well as their most recent or current locations. The classification of text based on priority enables efficient resource allocation during rescue operations. In conclusion, this system has the potential to reduce future flood-related fatalities. Additional research and development are necessary to thoroughly investigate this method’s practical capabilities and effectiveness.
