Tsunami Prediction using Random Forest

Among the most damaging natural catastrophes, tsunamis can cause great loss of life and widespread devastation of coastal regions. In disaster planning and risk reduction efforts, early tsunami warning and accurate forecast of events are essential. Our study introduces a Random Forest algorithm-based machine learning tsunami prediction system, a powerful ensemble learning method well known for its accuracy and robustness in handling complex data. Using well known earthquake characteristics like Magnitude, CDI (Community Decimal Intensity), MMI (Modified Mercalli Intensity), SIG (Significance), and NST (Number of Stations), the probability of a tsunami event (0: no tsunami, 1: tsunami) is found by means of training on an earthquake dataset. For better accessibility and ease of use, the system uses Django, a Python based web framework, to enable real time tsunami forecasting via a simple web interface. Processed by the system using the trained Random Forest model, the system can take seismic parameters from the users and deliver instantaneous forecasts. For disaster management officials, scientists, and the general public, the internet-based interface offers an easy and interactive experience supporting better accuracy and availability in tsunami threat evaluation. In this project, the application of web technologies and machine learning shows the feasibility of artificial intelligence early alert systems for anticipating natural catastrophes.