Novel Flood Prediction System based on Big Data and Machine Learning

The One of nature’s most destructive hazards, floods cause both human fatalities and property destruction. Numerous cities are influenced by the monsoon, therefore they frequently experience calamity. Early warning of a flood incident could help the public and the authorities plan both short-term and long-term preventive actions, prepare for evacuation and rescue efforts, and provide relief for flood victims. For example, one of the main factors in most flood management is the geographic locations of the affected areas and their respective severity. There is still no reliable method for predicting floods in advance. Existing technologies typically relied on prepared and manually entered data. Since the procedures were time-consuming, making real-time and early forecasts impossible. Additionally, these systems did not fully utilize the more thorough data that was accessible in the most recent big data platforms. As a result, this article suggests a unique flood forecasting method based on combining large data from crowd sources, geospatial systems, and meteorological systems. Modern learning strategies were at the heart of data intelligence. Both subjective and objective assessments showed that the proposed method could predict flood episodes that would occur in particular locations and times. Later benchmarking studies showed that the MLP ANN-equipped system provided the best accurate prediction, with correct percentage, Kappa, MAE, andRMSE values of 97.93, 0.89, 0.01, and 0.10, respectively