Advanced Machine Learning Methods for Major Hurricane Forecasting

Hurricanes, rapidly increasing in complexity and strength a warmer world, are one of the worst natural disasters 21st century. Further studies integrating changing hurricane features thus crucial to aid prediction major hurricanes. With this mind, we present new framework based on automated decision tree analysis, which has capability identify most important cloud structural parameters from GOES imagery as predictors for intensification potential Atlantic Pacific oceans. The proposed been proved effective predicting hurricanes with an overall accuracy 73% 6 54 h advance (both regions combined).