PROACTIVE DISASTER DETECTION

This paper introduces the Disaster Prediction Dashboard, a machine learning tool designed to predict the risks associated with hurricanes, earthquakes, and floods. It allows users to upload their own datasets, effectively managing missing values through imputation and enhancing accuracy with scaling techniques. By utilizing models such as random forest and logistic regression, it forecasts hazardous events based on various environmental factors, including wind speed, temperature, rainfall, and soil moisture. The methodology encompasses data preprocessing, feature selection, model training, and accuracy evaluation. Users can easily interact with a straightforward interface to upload data, enter specific values, and receive forecasts accompanied by clear risk indicators. With a focus on achieving high accuracy in risk classification, the system aims to improve disaster preparedness and aid in decision-making. Its scalability and user-friendly design highlight the successful application of machine learning in tackling environmental challenges.