Cyclone Intensity Estimation Using Machine Learning

Cyclones are one of the most devastating natural disasters, causing significant damage to life and property. Predicting cyclones with accuracy is crucial for mitigating their effects and preparing communities for potential impacts. This project aims to develop a machine learning-based cyclone prediction model using historical weather data, such as temperature, pressure, wind speed, and humidity. The model is trained on a synthetic dataset, as well as real-world data, to classify the likelihood of cyclone occurrences. The approach utilizes logistic regression as the primary classification algorithm, leveraging meteorological factors to predict the occurrence of cyclones (binary classification). The model is designed to handle both historical and realtime weather data, allowing for timely predictions of cyclone events. A graphical user interface (GUI) is incorporated into the system, providing an intuitive platform for users to input weather data and visualize predictions. This feature makes the tool accessible to meteorologists, disaster management authorities, and the general public, enabling informed decision-making and early warning systems. Through this project, we demonstrate how machine learning can enhance cyclone prediction accuracy and provide a reliable tool for disaster preparedness. The scalability of the model allows for future integration with real-time sensors, satellite data, and cloud computing platforms, paving the way for more robust and dynamic cyclone prediction systems.