Introduction to the Project

Tropical Cyclone is a low pressure system formed over tropical oceans. Warm, moist air over the ocean surface rises and condenses to release latent to energize the system, while cooler, denser, higher pressure air is attracted to the lower pressure water surface. Such that a convection is formed, together with the Coriolis force due to the Earth's rotation, a tropical cyclone is formed.

Tropical cyclone with high intensity is dangerous as it is causing damage to property, injury and even death. Thus it is necessary to have typhoon predictor which gives accurate prediction on the path, intensity of the incoming typhoons.

This project aims to create a track predictor model using machine learning techniques. It is supposed to train a model of neural network using large amount of historical track data which are extracted from several meteorological centers. The trained model is supposed to be able to give predictions of typhoon trajectory when its previous path is given as input.